Like I mentioned, I had to do three annotated bibliographies this semester. When I finished the one for my Feminist Theory class, it definitely helped with the project for that course, especially the analysis paper. Having that foundation made the writing process so much smoother.

This section explains how annotated bibliographies support deeper thinking rather than serving only as an organizational assignment. It shows how writing annotations can strengthen your analysis and help you develop clearer, more effective arguments throughout the research process.

This is true — this semester I had to complete three of them for different classes, which was definitely interesting. They’re not exactly my strong suit, but having this course really helped make the process easier, thankfully.

This was helpful when it came ot the note for assignments to make sure we are using the BEAM method.

I completely agree, and I’m glad you brought this up. The connection between keeping organized notes and avoiding plagiarism is so important in research — we never want to risk plagiarizing simply because our notes were unclear or incomplete. Staying organized really does protect both the quality and integrity of your work.

Very true — having a solid game plan makes the whole process flow more smoothly and feel far less stressful. It’s amazing how much easier research becomes when you know exactly what steps you’re taking and why.

James, thanks for the research and advertisements. Having just picked up one of these in lovely condition, I would generally agree with your assessment on the delineation of the two models: a "Report Electric" and a "Report de Luxe (SKE)".

Currently the TWdB has three different pages for what one might call the Olympia Report de Luxe (SKE) and which could be concatenated into a single model on one page:

1. "Olympia Report" https://typewriterdatabase.com/Olympia.Report.61.bmys with two exemplars from '75 and '78 which are explicitly badged as "Olympia Report de Luxe" on the hood

2. "Olympia Report deLuxe" https://typewriterdatabase.com/Olympia.Report+deLuxe.61.bmys which are all the badged the same, but somehow seem to have left the space between the "de" and "Luxe" out.

3. "Olympia SKE Report de Luxe" https://typewriterdatabase.com/Olympia.SKE+Report+de+Luxe.61.bmys which are all badged as "Report de Luxe" on the hood, but which include the SKE in the name because of the sticker on the side.

Personally, for ease of internet search most are likely to search for "Olympia Report de Luxe" though some may see the sticker near the power cord that reads "Typewriter Model SKE" (either on their physical machine or photos on eBay, Goodwill, etc.), so listing it in the database as "Olympia Report de Luxe (SKE)" may make the search most fruitful.

If the renaming of these three pages, which seem to be for a single model, does occur it would be useful to do a 403 redirect from the original pages to the final page so that the search engine optimization for these pages isn't lost. Adding a note to the model on the main Olympia page will help to clear up the details for future typewriter hunters as well.

Reply to https://typewriterdatabase.com/1973-olympia-report-electric.27118.typewriter

I agree this is an important reminder because citation software can sometimes create a false sense of accuracy. The section does a great job emphasizing that students still need to review their citations carefully instead of trusting the software automatically.

This section was helpful because many students are familiar with citation generators but may not realize there are tools designed to organize and manage sources for larger research projects—I definitely didn’t know that before. It’s a good reminder that research tools go beyond just creating citations and can actually make the whole process more manageable.

This stood out because bias is often easier to spot in sources we disagree with than in ones that align with our existing beliefs. The section does a great job showing that critical thinking means examining all sources carefully, even those that seem convincing at first glance

Very True—online info can look super polished and professional, even when it's misleading or just a joke. A lot of websites pretend to be reliable sources to push a point or sell something, and satire sites often mimic real news outlets. That's why it's important to take a close look at what the info is really trying to say before deciding to use it.

New doesn’t always mean better; we have to be careful to consider the quality of the material and the context it’s being used in.

Great point! This really connects source evaluation to information literacy as a whole. The section reinforces that research isn’t just about finding information—it’s also about deciding whether that information is trustworthy and genuinely useful.

Great tip! This stood out because it’s so different from how most people search on Google. The section explains really well how your search approach needs to shift depending on the type of tool you’re using, which makes the whole process feel more intentional and effective

The emphasis on database scope makes the research process feel more strategic and organized, which really helps with working more efficiently. Saving yourself time along the way can make the whole process much easier to manage.

This section was helpful because many students assume Google can find everything they need for academic research—especially since Google Scholar exists. The explanation reinforces why specialized databases are still essential for accessing scholarly and discipline‑specific materials that general search tools often miss

I thought this part and the comparison above were helpful is being able to tell if its a Database vs Catalog source.

Explaining this difference clearly helps reduce confusion when searching for sources.

I thought this was really helpful. Wikipedia is often treated as a “lazy” or unreliable shortcut, but this section explains why encyclopedias usually aren’t cited instead of just saying not to use them. It presents Wikipedia as a useful starting point for learning rather than something to dismiss entirely, which feels like a much more balanced approach

I agree that summarizing and paraphrasing are often treated like simple tasks, when they actually require a solid understanding of the material and careful interpretation.

Great tips!

Citation rules can definitely feel overwhelming, especially when you’re trying to figure out what counts as common knowledge or how to paraphrase correctly.

If we're going to do history, let's get it right. The Royal HH and REs offered color in the '52-54 range, but they were the exception rather than the rule. These are very difficult to find now. The Quiet De Luxes didn't get color until into 1956 and continued until 1958 when they were replaced by the (also) colorful Royal Futuras.

There were pockets of the late 20s and early 30s when Royal and Underwood among others experimented with color on portables as well, but these tended to be more basic reds, blues, and greens. The Corona 4s models were finished in DuPont DUCO® from 1927 to 1939. In a similar time period the Royal P also came in a small variety of colors.

The website utilizes images that helps people to visually understand the situation to support those who are not able to through reading long texts due to visual or cognitive challenges.

The webpage uses a big and bold headline with simple wording to clearly state the context of the news to help readers understand the issue quickly. This helps people with visual challenges to be able to identify the issue in a clear and easy way.

The webpage is not fully understandable. There are a lot of advertisements on this page. Having a lot of advertisements can clutter and distract readers, making it challenging to focus on the news. The webpage contains extra content that would overstimulate and it distract audiences from the main content.

The website is operable for people living with motor impairment because it does not contain features such as hover interactions that require a mouse to interact with the webpage.

The website demonstrates good robust and accessibility because the webpage allows users to navigate the menu easily utilizing their assistive devices such as a keyboard or screen readers.

The reminder to check assignment instructions carefully is helpful since students sometimes assume that one citation style works for every course, especially if they are in different academic style classes (General vs. Social Science).

Good tip- Keeping citation information organized early seems especially important when working with large numbers of sources.

The discussion about shifting online habits versus academic expectations was interesting because so many people are used to informal linking and sharing online. This section clearly explains why formal citation practices still matter in academic writing, even as technology and online behavior keep changing.

I liked how this was pointed out, mentioning that it’s totally normal not to get all the scholarly stuff and that sometimes you need to do a bit more research.

This is very important to be aware of which citation style has been used; switching between APA, MLA, and other citation styles can become confusing very quickly, especially when taking courses from different subject areas at the same time.

The advice about keeping track of citation information early is practical and could prevent a lot of last-minute stress.

Duplicate paragraph already stated above; I don't know if this is a typo but something that should be noted.

eLife Assessment

This important study fills a major geographic and temporal gap in understanding Paleocene mammal evolution in Asia and proposes an intriguing "brawn before bite" hypothesis grounded in diverse analytical approaches. The work rests on a solid methodological base. Some limitations remain, including uncertainty introduced by pooling different tooth positions, limited dietary interpretation, and the predominantly herbivorous taxonomic focus, which narrows the ecological scope of the conclusions. However, the manuscript provides a substantially strengthened and well-supported contribution, while appropriately inviting further work to clarify dietary trends, broader ecological context, and links between dental trait evolution and environmental change.

Reviewer #2 (Public review):

Summary:

This study uses dental traits of a large sample of Chinese mammals to tract evolutionary patterns through the Paleocene. It presents and argues for a 'brawn before bite' hypothesis -- mammals increased in body size disparity before evolving more specialized or adapted dentitions. The study makes use of an impressive array of analyses, including dental topographic, finite element, and integration analyses, which help to provide a unique insight into mammalian evolutionary patterns.

Strengths:

This paper helps to fill in a major gap in our knowledge of Paleocene mammal patterns in Asia, which is especially important because of the diversification of placentals at that time. The total sample of teeth is impressive and required considerable effort for scanning and analyzing. And there is a wealth of results for DTA, FEA, and integration analyses. Further, some of the results are especially interesting, such as the novel 'brawn before bite' hypothesis and the possible link between shifts in dental traits and arid environments in the Late Paleocene. Overall, I enjoyed reading the paper and I think the results will be of interest to a broad audience.

Weaknesses:

For the original draft of the manuscript, I had four major concerns with the study, especially related to the sampling, diet, and evidence for the 'brawn before bite' hypothesis. I still believe that the original issues that I raised may be weaknesses of the study. For example, there is still limited discussion on diets (even though the dental topographic analyses used in the study are designed for inferring diets). And I find the results a little challenging to interpret because teeth of multiple positions are included in the same samples, which seems problematic. That said, the authors have addressed each of my previous concerns and have made major revisions, including running new analyses, and thus I support the paper.

Author response:

The following is the authors’ response to the original reviews

eLife Assessment

This important study fills a major geographic and temporal gap in understanding Paleocene mammal evolution in Asia and proposes an intriguing "brawn before bite" hypothesis grounded in diverse analytical approaches. However, the findings are incomplete because limitations in sampling design - such as the use of worn or damaged teeth, the pooling of different tooth positions, and the lack of independence among teeth from the same individuals - introduce uncertainties that weaken support for the reported disparity patterns. The taxonomic focus on predominantly herbivorous clades also narrows the ecological scope of the results. Clarifying methodological choices, expanding the ecological context, and tempering evolutionary interpretations would substantially strengthen the study.

We have now thoroughly revised our manuscript in response to the editor and reviewer’s comments. In particular with regard to:

(1) Sampling design: we clarified our methods section to indicate that we did not use worn or broken teeth in our initial analyses. We added the following sentence around line 690:

“These tooth positions were selected from a broader examination of ~300 individual teeth from 72 specimens. We vetted the specimens and excluded 99 tooth positions (~33% of teeth initially chosen for possible inclusion) from our analyses because they either (1) were partially or completely broken at the crown, (2) were in an advanced stage of attritional wear where no cusps could be identified, or (3) possessed a combination of the two aforementioned conditions.”

(2) Pooled versus by-tooth position analyses: we repeated the three major analyses (DTA & FEA variability through time, tooth size and variability through time, and DTA-FEA correlation through time) for individual molars (upper M1-3, lower m1-3) and select premolars (upper P3-P4 and lower p4; lower and upper p2 samples contained fewer than 5 specimens across the three time intervals, lower p3 contained only 2 specimens for the middle Paleocene, so they were excluded from the sub-partition analyses).

For DTA & FEA variability through time (summarized as a new figure, Fig. S5, also pasted below), OPCR, DNE, and FEA trait data are supported in 78-100% of the per-tooth analyses for both the early-middle Paleocene and middle-late comparisons. By contrast, RFI and Slope data are replicated in only 22-56% of the per-tooth analyses. We qualified the main text reporting and discussion to include these sensitivity analyses so readers can assess nuances in the data when comparing pooled sample versus per-tooth analyses.

For tooth size and variability through time (summarized in a new table, Table S3, also pasted below), we observed broad concordance in the pooled analyses and the per-tooth partitioned analyses. Different tooth positions provide strong support for different aspects of the observed trends, with the lower fourth premolar being the strongest driver of the overall trend. All of the significant trends in per-tooth analyses are in the same direction (i.e., decreasing size disparity and size mean through time) as the pooled sample. We added qualifying clarification in the text to bring attention to these refined results.

For DTA-FEA correlation through time, we generated per-tooth correlation plots in three new figures (Figs. S9-11, only Fig. S10 shown here as an example). We observed that upper M1 patterns general reflect the trend recovered from analysis of the overall dataset, but M2 and M3 results display inconsistent DTA-FEA correlations, possibly due to small sample sizes. Lower molar patterns generally replicate those recovered in the overall analyses, but lower M1 and M2 signals appear to be stronger than those for lower M3. Finally, low sample sizes make premolar correlations unstable, with general pattern showing EP-MP strengthening then MP-LP stasis or weakening. Given these findings, it appears that the results in the pooled sample correlation plots are mainly driven by lower molar signals. It is not possible to conclude the other tooth position display different patterns because of the limited sample sizes.

(3) Ecological scope of the study: although carnivorans and mesonychids are recorded from some of the time intervals examined in this study, our sampling choice of pantodonts and anagalids reflects the high abundance of available dental specimens in those clades, permitting us to make the strongest statistical inference given the incomplete fossil record. Additionally, all sampled taxa come from archaic clades that have not been determined to be specifically herbivorous; we included an additional paragraph in the introduction to explain this:

“A major challenge with expanding analyses of post K-Pg recovery to Paleocene mammal assemblages elsewhere in the world is the generally stratigraphically limited nature of early Cenozoic sequences. In Asia, Paleocene localities in China represent the best studied to date[11]. From the earliest Paleocene, highly regional and endemic faunas are known from a handful of sedimentary basins (Fig. S1A). Among the faunal elements, only the archaic clades Anagalida and Pantodonta are consistently sampled across the major subdivisions of the Paleocene[11]. An additional complication with ecomorphological analysis of these early mammals is the uncertainty in their dietary ecology, as they are beyond the reach of conventional phylogenetic bracketing approaches to dietary reconstruction. Phenomic analysis of the placental radiation supports insectivory as the ancestral diet of the hypothetical placental ancestor, but uncertainty in the post K-Pg availability of insects and plants in some regions leave some doubt as to the accuracy of this ancestral state reconstruction[1]. Herein we treat the archaic Paleocene taxa in our analyses as having generalized diets rather than categorizing them as insectivores, herbivores, or carnivores.”

Public Reviews:

Reviewer #1 (Public review):

Summary:

This work provides valuable new insights into the Paleocene Asian mammal recovery and diversification dynamics during the first ten million years post-dinosaur extinction. Studies that have examined the mammalian recovery and diversification post-dinosaur extinction have primarily focused on the North American mammal fossil record, and it's unclear if patterns documented in North America are characteristic of global patterns. This study examines dietary metrics of Paleocene Asian mammals and found that there is a body size disparity increase before dietary niche expansion and that dietary metrics track climatic and paleobotanical trends of Asia during the first 10 million years after the dinosaur extinction.

Strengths:

The Asian Paleocene mammal fossil record is greatly understudied, and this work begins to fill important gaps. In particular, the use of interdisciplinary data (i.e., climatic and paleobotanical) is really interesting in conjunction with observed dietary metric trends.

Weaknesses:

While this work has the potential to be exciting and contribute greatly to our understanding of mammalian evolution during the first 10 million years post-dinosaur extinction, the major weakness is in the dental topographic analysis (DTA) dataset.

There are several specimens in Figure 1 that have broken cusps, deep wear facets, and general abrasion. Thus, any values generated from DTA are not accurate and cannot be used to support their claims. Furthermore, the authors analyze all tooth positions at once, which makes this study seem comprehensive (200 individual teeth), but it's unclear what sort of noise this introduces to the study. Typically, DTA studies will analyze a singular tooth position (e.g., Pampush et al. 2018 Biol. J. Linn. Soc.), allowing for more meaningful comparisons and an understanding of what value differences mean. Even so, the dataset consists of only 48 specimens. This means that even if all the specimens were pristinely preserved and generated DTA values could be trusted, it's still only 48 specimens (representing 4 different clades) to capture patterns across 10 million years. For example, the authors note that their results show an increase in OPCR and DNE values from the middle to the late Paleocene in pantodonts. However, if a singular tooth position is analyzed, such as the lower second molar, the middle and late Paleocene partitions are only represented by a singular specimen each. With a sample size this small, it's unlikely that the authors are capturing real trends, which makes the claims of this study highly questionable.

With regard to sampling design: we clarified our methods section to indicate that we did not use worn or broken teeth in our initial analyses. We added the following sentence around line 690:

“These tooth positions were selected from a broader examination of ~300 individual teeth from 72 specimens. We vetted the specimens and excluded 99 tooth positions (~33% of teeth initially chosen for possible inclusion) from our analyses because they either (1) were partially or completely broken at the crown, (2) were in an advanced stage of attritional wear where no cusps could be identified, or (3) possessed a combination of the two aforementioned conditions.”

With regard to pooled versus by-tooth position analyses: we repeated the three major analyses (DTA & FEA variability through time, tooth size and variability through time, and DTA-FEA correlation through time) for individual molars (upper M1-3, lower m1-3) and select premolars (upper P3-P4 and lower p4; lower and upper p2 samples contained fewer than 5 specimens across the three time intervals, lower p3 contained only 2 specimens for the middle Paleocene, so they were excluded from the sub-partition analyses).

For DTA & FEA variability through time (summarized as a new figure, Fig. S5, also pasted below), OPCR, DNE, and FEA trait data are supported in 78-100% of the per-tooth analyses for both the early-middle Paleocene and middle-late comparisons. By contrast, RFI and Slope data are replicated in only 22-56% of the per-tooth analyses. We qualified the main text reporting and discussion to include these sensitivity analyses so readers can assess nuances in the data when comparing pooled sample versus per-tooth analyses.

For the tooth size and variability through time (summarized in a new table, Table S3, also pasted below), we observed broad concordance in the pooled analyses and the per-tooth partitioned analyses. Different tooth positions provide strong support for different aspects of the observed trends, with the lower fourth premolar being the strongest driver of the overall trend. All of the significant trends in per-tooth analyses are in the same direction (i.e., decreasing size disparity and size mean through time) as the pooled sample. We added qualifying clarification in the text to bring attention to these refined results.

For DTA-FEA correlation through time, we generated per-tooth correlation plots in three new figures (Figs. S8-10, only Fig. S9 shown here as an example). We observed that upper M1 patterns general reflect the trend recovered from analysis of the overall dataset, but M2 and M3 results display inconsistent DTA-FEA correlations, possibly due to small sample sizes. Lower molar patterns generally replicate those recovered in the overall analyses, but lower M1 and M2 signals appear to be stronger than those for lower M3. Finally, low sample sizes make premolar correlations unstable, with general pattern showing EP-MP strengthening then MP-LP stasis or weakening. Given these findings, it appears that the results in the pooled sample correlation plots are mainly driven by lower molar signals. It is not possible to conclude the other tooth position display different patterns because of the limited sample sizes.

Reviewer #2 (Public review):

Summary:

This study uses dental traits of a large sample of Chinese mammals to track evolutionary patterns through the Paleocene. It presents and argues for a 'brawn before bite' hypothesis - mammals increased in body size disparity before evolving more specialized or adapted dentitions. The study makes use of an impressive array of analyses, including dental topographic, finite element, and integration analyses, which help to provide a unique insight into mammalian evolutionary patterns.

Strengths:

This paper helps to fill in a major gap in our knowledge of Paleocene mammal patterns in Asia, which is especially important because of the diversification of placentals at that time. The total sample of teeth is impressive and required considerable effort for scanning and analyzing. And there is a wealth of results for DTA, FEA, and integration analyses. Further, some of the results are especially interesting, such as the novel 'brawn before bite' hypothesis and the possible link between shifts in dental traits and arid environments in the Late Paleocene. Overall, I enjoyed reading the paper, and I think the results will be of interest to a broad audience.

Weaknesses:

I have four major concerns with the study, especially related to the sampling of teeth and taxa, that I discuss in more detail below. Due to these issues, I believe that the study is incomplete in its support of the 'brawn before bite' hypothesis. Although my concerns are significant, many of them can be addressed with some simple updates/revisions to analyses or text, and I try to provide constructive advice throughout my review.

(1) If I understand correctly, teeth of different tooth positions (e.g., premolars and molars), and those from the same specimen, are lumped into the same analyses. And unless I missed it, no justification is given for these methodological choices (besides testing for differences in proportions of tooth positions per time bin; L902). I think this creates some major statistical concerns. For example, DTA values for premolars and molars aren't directly comparable (I don't think?) because they have different functions (e.g., greater grinding function for molars). My recommendation is to perform different disparity-through-time analyses for each tooth position, assuming the sample sizes are big enough per time bin. Or, if the authors maintain their current methods/results, they should provide justification in the main text for that choice.

With regard to pooled versus by-tooth position analyses: we repeated the three major analyses (DTA & FEA variability through time, tooth size and variability through time, and DTA-FEA correlation through time) for individual molars (upper M1-3, lower m1-3) and select premolars (upper P3-P4 and lower p4; lower and upper p2 samples contained fewer than 5 specimens across the three time intervals, lower p3 contained only 2 specimens for the middle Paleocene, so they were excluded from the sub-partition analyses).

For DTA & FEA variability through time (summarized as a new figure, Fig. S5, also pasted below), OPCR, DNE, and FEA trait data are supported in 78-100% of the per-tooth analyses for both the early-middle Paleocene and middle-late comparisons. By contrast, RFI and Slope data are replicated in only 22-56% of the per-tooth analyses. We qualified the main text reporting and discussion to include these sensitivity analyses so readers can assess nuances in the data when comparing pooled sample versus per-tooth analyses.

For the tooth size and variability through time (summarized in a new table, Table S3, also pasted below), we observed broad concordance in the pooled analyses and the per-tooth partitioned analyses. Different tooth positions provide strong support for different aspects of the observed trends, with the lower fourth premolar being the strongest driver of the overall trend. All of the significant trends in per-tooth analyses are in the same direction (i.e., decreasing size disparity and size mean through time) as the pooled sample. We added qualifying clarification in the text to bring attention to these refined results.

For DTA-FEA correlation through time, we generated per-tooth correlation plots in three new figures (Figs. S8-10, only Fig. S9 shown here as an example). We observed that upper M1 patterns general reflect the trend recovered from analysis of the overall dataset, but M2 and M3 results display inconsistent DTA-FEA correlations, possibly due to small sample sizes. Lower molar patterns generally replicate those recovered in the overall analyses, but lower M1 and M2 signals appear to be stronger than those for lower M3. Finally, low sample sizes make premolar correlations unstable, with general pattern showing EP-MP strengthening then MP-LP stasis or weakening. Given these findings, it appears that the results in the pooled sample correlation plots are mainly driven by lower molar signals. It is not possible to conclude the other tooth position display different patterns because of the limited sample sizes.

Also, I think lumping teeth from the same specimen into your analyses creates a major statistical concern because the observations aren't independent. In other words, the teeth of the same individual should have relatively similar DTA values, which can greatly bias your results. This is essentially the same issue as phylogenetic non-independence, but taken to a much greater extreme.

It seems like it'd be much more appropriate to perform specimen-level analyses (e.g., Wilson 2013) or species-level analyses (e.g., Grossnickle & Newham 2016) and report those results in the main text. If the authors believe that their methods are justified, then they should explain this in the text.

Based on the per-tooth partition analyses we performed and reported above, the results now show that the overall trends described in the previous draft of the study is a composite of signals from different regions of the dentition. For example, the OPCR, DNE, and FEA trends persist across most tooth positions, whereas the Slope and RFI trends are mainly driven by lower fourth premolar patterns. The tooth size results are also mainly driven by lower fourth premolar patterns, but tooth disparity trends are broadly supported across tooth positions. These observations indicate that the overall trends remain valid, but there are nuances as to which tooth positions are driving which components of the trends. As such, we deem the overall results to be valid, and focused our revision on providing the nuances so readers can assess through-time patterns in more detail than in the previous version of the study.

(2) Maybe I misunderstood, but it sounds like the sampling is almost exclusively clades that are primarily herbivorous/omnivorous (Pantodonta, Arctostylopida, Anagalida, and maybe Tillodonta), which means that the full ecomorphological diversity of the time bins is not being sampled (e.g., insectivores aren't fully sampled). Similarly, the authors say that they "focused sampling" on those major clades and "Additional data were collected on other clades ... opportunistically" (L628). If they favored sampling of specific clades, then doesn't that also bias their results?

If the study is primarily focused on a few herbivorous clades, then the Introduction should be reframed to reflect this. You could explain that you're specifically tracking herbivore patterns after the K-Pg.

We appreciate the reviewer’s suggestion that our sampling may have focused on putative herbivorous clades more than others. However, at the early stage of placental evolution during the Paleocene, and in particular among the endemic forms we studied from south China, it is unclear to us that such clearcut ecomorphological categories were present amongst the fossil mammals. Thus, we take a more agnostic approach and do not define the dietary categories of the sample taxa (and by extension, those of the unsampled taxa). Although we recognize that representatives of certain clades, such as Carnivora, may be more reasonably interpreted as carnivores/insectivores/omnivores and, in the current context, remains unsampled, we point out the fact that including tooth samples from rare taxa such as carnivores likely would have biased the analyses temporally. Chinese Paleocene carnivores are known only from one of the three time intervals analyzed (representing only a handful of specimens), and so would potentially inflate the disparity in that time interval relative to the others (if dentitions specialized for carnivory is assumed to be present in the Paleocene). To clarify this point, we added a paragraph in the introduction:

“A major challenge with expanding analyses of post K-Pg recovery to Paleocene mammal assemblages elsewhere in the world is the generally stratigraphically limited nature of early Cenozoic sequences. In Asia, Paleocene localities in China represent the best studied to date[11]. From the earliest Paleocene, highly regional and endemic faunas are known from a handful of sedimentary basins (Fig. S1A). Among the faunal elements, only the archaic clades Anagalida and Pantodonta are consistently sampled across the major subdivisions of the Paleocene[11]. An additional complication with ecomorphological analysis of these early mammals is the uncertainty in their dietary ecology, as they are beyond the reach of conventional phylogenetic bracketing approaches to dietary reconstruction. Phenomic analysis of the placental radiation supports insectivory as the ancestral diet of the hypothetical placental ancestor, but uncertainty in the post K-Pg availability of insects and plants in some regions leave some doubt as to the accuracy of this ancestral state reconstruction[1]. Herein we treat the archaic Paleocene taxa in our analyses as having generalized diets rather than categorizing them as insectivores, herbivores, or carnivores.”

(3) There are a lot of topics lacking background information, which makes the paper challenging to read for non-experts. Maybe the authors are hindered by a short word limit. But if they can expand their main text, then I strongly recommend the following:

a) The authors should discuss diets. Much of the data are diet correlates (DTA values), but diets are almost never mentioned, except in the Methods. For example, the authors say: "An overall shift towards increased dental topographic trait magnitudes ..." (L137). Does that mean there was a shift toward increased herbivory? If so, why not mention the dietary shift? And if most of the sampled taxa are herbivores (see above comment), then shouldn't herbivory be a focal point of the paper?

We edited the introduction to say that “We used dental topographical traits as indicators of ecomorphological diversity[28] and examined temporal shifts in tooth crown complexity, curvature, and height and their association with tooth performance in terms of deformation resistance using topographic and simulation analyses.” And also added the following to the methods section, in order to clarify that we are using DTA as a general ecomorphological proxy, and not a direct dietary proxy.

“Overall, we use these DTA traits as indicators of ecomorphological capacity, but do not link them explicitly to dietary categories. The craniodental morphology of archaic placental clades in general have not been demonstrated to share the same structure-function linkages as crown mammals, so the aforementioned linkages between DTA and dietary ecology in extant species only serve as evidence that DTA is a potentially useful ecomorphological proxy, without the application of those DTA-diet relationships to the Paleocene fossil mammal dataset.”

b) The authors should expand on "we used dentitions as ecological indicators" (L75). For non-experts, how/why are dentitions linked to ecology? And, again, why not mention diet? A strong link between tooth shape and diet is a critical assumption here (and one I'm sure that all mammalogists agree with), but the authors don't provide justification (at least in the Introduction) for that assumption. Many relevant papers cited later in the Methods could be cited in the Introduction (e.g., Evans et al. 2007).

We added the following sentence to clarify our usage of tooth crowns as ecomorphological proxies: “Teeth are among the most well-preserved parts of fossil mammals, and the fact that they interface directly with the environment through mastication makes them suitable elements for studying potential ecology-morphology linkages.”

c) Include a better introduction of the sample, such as explicitly stating that your sample only includes placentals (assuming that's the case) and is focused on three major clades. Are non-placentals like multituberculates or stem placentals/eutherians found at Chinese Paleocene fossil localities and not sampled in the study, or are they absent in the sampled area?

We modified the following sentence to indicate our sampling focus on placentals: “Our analyses focused on placental mammals from three of the most fossiliferous and biogeographically isolated Paleocene sedimentary sequences in paleotropical Asia: The Nanxiong, Qianshan, and Chijiang Basins in present-day south China 23–27 (Fig. S1)”

d) The way in which "integration" is being used should be defined. That is a loaded term which has been defined in different ways. I also recommend providing more explanation on the integration analyses and what the results mean.

If the authors don't have space to expand the main text, then they should at least expand on the topics in the supplement, with appropriate citations to the supplement in the main text.

We replaced all mentions of “integration” with “covariation” to avoid using the loaded terminology. Covariation more accurately reflects the correlation between two sets of traits (DTA vs FEA) without invoking developmental mechanisms implied by modularity/integration.

(4) Finally, I'm not convinced that the results fully support the 'brawn before bite' hypothesis. I like the hypothesis. However, the 'brawn before ...' part of the hypothesis assumes that body size disparity (L63) increased first, and I don't think that pattern is ever shown. First, body size disparity is never reported or plotted (at least that I could find) - the authors just show the violin plots of the body sizes (Figures 1B, S6A). Second, the authors don't show evidence of an actual increase in body size disparity. Instead, they seem to assume that there was a rapid diversification in the earliest Paleocene, and thus the early Paleocene bin has already "reached maximum saturation" (L148). But what if the body size disparity in the latest Cretaceous was the same as that in the Paleocene? (Although that's unlikely, note that papers like Clauset & Redner 2009 and Grossnickle & Newham 2016 found evidence of greater body size disparity in the latest Cretaceous than is commonly recognized.) Similarly, what if body size disparity increased rapidly in the Eocene? Wouldn't that suggest a 'BITE before brawn' hypothesis? So, without showing when an increase in body size diversity occurred, I don't think that the authors can make a strong argument for 'brawn before [insert any trait]".

Although it's probably well beyond the scope of the study to add Cretaceous or Eocene data, the authors could at least review literature on body size patterns during those times to provide greater evidence for an earliest Paleocene increase in size disparity.

We added a sentence in the discussion of body size during the Paleocene to note that the largest late Cretaceous fossil mammals in China are shrew- to gopher-sized, whereas the largest early Paleocene Chinese Endemic Pantodonts are dog-sized:

“Dog-sized CEPs such as Bemalambda reached sizes not seen in late Cretaceous mammals from China such as Zhangolestes and Kryptobaatar, which are shrew- to gopher-sized [Meng 2014]”

Reference: Meng, J. (2014). Mesozoic mammals of China: implications for phylogeny and early evolution of mammals. Natl. Sci. Rev. 1, 521–542. 10.1093/nsr/nwu070.

Furthermore, we tempered our discussion to restrict the “brawn before bite” hypothesis to post K-Pg recovery in the Paleocene. Body size patterns shifted in the Eocene as crown clades replaced the archaic endemic clades analyzed in our study, and much larger taxa began to appear after the PETM. Such body size shift patterns are based on different clades and likely different dynamics compared to the 10-million year interval examined in our study, so we refrain from commenting on post-Paleocene times.

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors):

(1) In regard to the DTA dataset: Was there a method used to 'fix' these teeth before dental topographic analyses were implemented? If so, this should be explicitly stated. If not, the authors should explain why broken, worn, or abraded teeth were used.

We excluded the incomplete teeth from our analyses. We added the following sentence for clarification: “These tooth positions were selected from a broader examination of ~300 individual teeth from 72 specimens. We vetted the specimens and excluded 99 tooth positions (~33% of teeth initially chosen for possible inclusion) from our analyses because they either (1) were partially or completely broken at the crown, (2) were in an advanced stage of attritional wear where no cusps could be identified, or (3) possessed a combination of the two aforementioned conditions.”

(2) The authors should explicitly explain why all tooth positions were analyzed together. Again, this is not something that is typically done, and some explanation would be helpful for readers.

We added a paragraph in the methods section to explain both our pooled sampling approach, as well as the per-tooth analyses added in this revised manuscript:

“Given the rarity of Paleocene fossil material from China, we combined data from different tooth positions into three pooled samples, one for each of the time intervals examined (early, middle, late Paleocene). We treated the pooled samples as representative of the range of dental topographic features and bite performance traits available to the mammal taxa under study. In this way, the variance estimates are interpreted as measures of the morphological and performance heterogeneity present in each time interval dataset. To further tease out the possibility of specific tooth positions driving the overall trends observed in the pooled samples, we also performed the DTA, FEA, DTA-FEA correlation, and tooth size through-time analyses using per-tooth data partitions.”

(3) I think the authors should hedge their claims a bit more and recognize the limitations of their study (e.g., sample size and tooth preservation).

We thank the reviewer for raising this important point. We carefully read through the main text and further tempered our interpretations based on the limitations of our data. Additionally, we added a paragraph in the supplemental text to summarize the major sources of uncertainty in the sample:

“Sample and methodological limitations

The highly fragmentary nature of early Cenozoic mammal fossils in Asia means that even the best preserved faunas studied herein contain much missing information. First, the absence of a high-resolution chronological framework prevents the fossil data from being analyzed on a continuous time axis; the binning of the samples into three main intervals within a 10-million-year period hinders additional hypotheses about the environmental and climatic correlations of the dental structure-performance results presented. Second, the uneven sampling of the available mammalian assemblage throughout the Paleocene sites in China limits the breadth of ecomorphological categories included in the analyses; rarer taxa representing more specialized carnivore, insectivore, or herbivore forms were not included in our sampling. Third, the spatial discontinuity of stratigraphically younger (Eocene) and older (Cretaceous) mammal assemblages means that body size and ecomorphological shifts bracketing the Paleocene cannot currently be analyzed alongside the dataset presented. These limitations should be taken into account when considering the interpretations made in the main text.”

Reviewer #2 (Recommendations for the authors):

I'm including my Line Comments here as recommendations for the authors. But note that many of my recommendations are also in my Public Review.

L22: "3% of sites"? Do you mean 3% of global sites?

Yes, we revised the sentence to indicate 3% of global sites. Thank you for this suggestion.

L35: This is nitpicky because it's not crucial to your study, but I can't help but point out that the Long Fuse, etc, hypotheses are specifically about the DIVERGENCE TIMES for Placentalia and major subclades, NOT the 'adaptive radiation' of placentals like you imply in your text. Adaptive radiations include ecomorphological diversification and are driven by ecological opportunity (e.g., Schluter 2000). (Emphasis on 'ecological.') The long fuse, short fuse, and explosive models do not include an ecological component - i.e., the diversifications could have occurred without ecological diversification. Instead, for hypotheses that are specifically on the adaptive/ecological radiation of mammals, see the Early Rise, Suppression (or Dinosaur Incumbency; Benevento et al. 2023 Palaeontology), and Late Rise hypotheses (Grossnickle et al. 2019 TREE). These hypotheses apply broadly to all mammals, not just placentals (see Box 1's figure in Grossnickle et al. 2019), but they can still be applied to mammalian subclades like eutherians/placentals (e.g., see Thomas Halliday papers).

Thank you for helping to clarify the adaptive radiation vs. divergence time concepts. We edited this sentence to mention the adaptive radiation hypotheses instead, adding in the references provided by the reviewer.

L39-40: I think your comment is probably accurate. But keep in mind that advocates of the Early Rise and Delayed Rise hypotheses (see citations within Grossnickle et al. 2019) might argue that other time periods, other than the Paleocene, are equally or more important.

We added a reference to Grossnickle et al. 2019 to bring attention to potential arguments otherwise. Thank you for the suggestion.

L48: I think the inclusion of "at higher latitudes" is a little distracting or misleading and should be erased. It implies that the taxonomic diversification was ONLY rapid at higher latitudes. But many of the references that you cite include analyses at the global or continental scale (e.g., Alroy 1999, Grossnickle & Newham 2016) and don't distinguish patterns at different latitudes. If you want to keep the point about latitudes, then I recommend inserting a separate sentence on that point.

We removed “at higher latitudes”.

L50: Isn't "stem lineages and those with no living relatives" somewhat redundant? Or do you mean something like "stem placental/eutherian lineages and extinct placental subgroups"?

Yes, we adopted the suggested phrasing. Thank you.

L53: I recommend starting a new paragraph around here (maybe starting with "Distinct from ...") that focuses specifically on introducing the 'brawn before [ecomorphological trait]' hypothesis.

Done.

L56: "large herbivores and their predators"? Are you just referring to mammals? Wilson (2013), which you cite, and Grossnickle & Newham (2016) argued that dietary specialists were targeted at the K-Pg, but none of the herbivores were "large" (at least relative to Cenozoic herbivores). And most faunivorous mammals at the time were probably insectivorous and not preying on herbivorous mammals, besides maybe a few outlying taxa (e.g., Altacreodus, Nanocuris). I'd revise your sentence for clarity.

We removed “disproportionately impacting large herbivores and their predators” for clarity.

L63: I'd replace "ecometric" with "ecomorphological". Ecometrics commonly refers to using fossil traits to infer paleo environments/climate (e.g., see papers by David Polly, Michelle Lawing, etc), which I don't think is what you're referring to here. (E.g., I don't think that brain size or jaw shape patterns were/are used to infer paleo environments.)

Revised. Thank you.

L85: I strongly advise against making conclusions like this: "Dental height and sharpness variability ... [spiked] in the middle Paleocene corresponding to a short-lived negative excursion in global temperature." That implies that the change in dentitions is linked to global temperature changes, which I don't think your results support. Later in the text you highlight the temporal uncertainty of your time bin ages (L650) and say that the middle Paleocene bin could be as old as ~62 Ma (L646), which is well before the negative excursion (and looks to be more in line with a positive excursion!), at least according to the Figure 1 time scale (see comment below). So, I don't think that your results even support your statement.

We reworded this sentence to say “Dental height and sharpness variability were low in the beginning and end of the time interval, with a peak in the middle Paleocene. This pattern is observed both when dentitions are considered holistically and by tooth position in the lower dentition (Fig. S5; upper teeth display the opposite pattern).”

L144: Using variance for disparity seems fine. But keep in mind that other disparity metrics, such as range (or sum-of-ranges for multivariate data), might produce different results. For instance, variance of RFI and Slope spike in the middle Paleocene, like you point out, but based on the values in Figure 1A, it looks like the ranges stay relatively constant through the Paleocene (although I realize that the ranges might change with bootstrapping). So, your choice of disparity metric might have a big influence on your conclusions. Alternatively, you could calculate disparity using multiple metrics (e.g., Brusatte et al. 2012 Nature Communications; Grossnickle & Newham 2016 supplemental analyses), even if it's just for supplemental analyses.

Thank you for bringing the choice of disparity measures to our attention. We conducted a parallel set of bootstrapped disparity calculation and comparison analyses using range lengths (maximum trait value – minimum trait value for a given trait) and summarized the through-time trends as for variance-based results (Fig. S5). Overall, very similar trends are observed, providing support for the variance-based data interpretation presented in the main text. We added explanation of this additional sensitivity testing both in the main text and in the supplemental text.

L147: "body size disparity ... (Fig. 1B, S6A, Table 1, Data S5)." But I don't see disparity calculated or plotted in any of the figures/tables that you cite. You test for differences in disparity between time bins (Table 1), but that doesn't provide the actual disparity patterns.

We generated a new figure (Fig. S8) to show the tooth size variance and range levels across time and data partitions, and modified this sentence to say that “Over the same time interval examined, body size disparity and mean were higher in the early Paleocene than in subsequent time intervals (Fig. S8, Table S3; also supported by premolar 4 and upper molar partition analyses), indicating that substantial increases in the disparity of dental complexity, curvature, and height lagged behind maximum size disparity tooth size during the Paleocene.”

L151-153: Maybe. But you're basing this on a much narrower temporal range (Paleocene) than the brain and jaw studies, and I think those studies observed big increases in brain/jaw disparity in the Eocene, which you don't sample. And as I explained elsewhere, I'm not convinced that your results strongly support the same pattern. At a minimum, I recommend tempering your conclusions to better reflect the uncertainty of your results.

We tempered our statements here to say that “This suggests a ‘brawn before bite’ pattern in endemic Asian mammals, partially mirroring the endocranial and jaw functional morphology patterns identified in their North American and European counterparts [21,22]. These findings raise the possibility that an initial size-driven post-K-Pg recovery followed by ecomorphological radiation was a global phenomenon, even as regional tectonic events such as the initial collision of the Indian subcontinent with Asia and Deccan Traps volcanism influenced local mammal evolution.”

L170: I'm not well-versed in integration (and modularity) studies, so maybe this reflects my ignorance, but I had trouble understanding sentences like this: "These findings indicate that form-function malleability, the coexistence of distinct topography-performance relationships in each time and taxon partition while overall integration between the two trait groups increases between time bins, was present throughout the Paleocene." If there is space, I recommend revising and/or breaking apart long, jargon-y sentences like that (throughout the paper) so that they're more digestible for readers.

We simplified complex sentences such as the one the reviewer noted, in order to communicate our findings and interpretations more clearly. Thank you for the suggestion.

L183: It's probably fine to assume most placental orders arose in the Paleocene based on fossil evidence. But keep in mind that molecular studies often argue that many orders arose in the Late Cretaceous.

We revised the statement to indicate a “Cretaceous/Paleocene” origin of many modern mammal orders.

L200-207: Again, this might just reflect my ignorance concerning integration analyses, but I recommend expanding on this text to better explain how your integration results support this conclusion. It seems really interesting, and I like the Garden of Eden hypothesis. It's just not immediately clear to me how your results support that hypothesis. A little more background on how to interpret the integration results would be helpful.

We expanded the discussion here to say that “Such flexibility in dental form-function linkage permits ‘mix and match’ trait combinations rather than evolutionary change as a single unit, potentially enhancing the evolvability of feeding ecological traits as new environmental conditions arose [Goswami et al. 2015]”

Reference: Goswami, A., Binder, W.J., Meachen, J., and O’Keefe, F.R. (2015). The fossil record of phenotypic integration and modularity: A deep-time perspective on developmental and evolutionary dynamics. Proc. Natl. Acad. Sci. 112, 4891–4896. 10.1073/pnas.1403667112.

L218: "reached maximum tooth size disparity early". Again, I don't see size disparity plotted or reported. And without baseline comparisons (Late K or Eocene), it's hard to interpret your results and evaluate what 'maximum' means (Figure 1B).

We revised the sentence to now say “In response, Paleocene mammal clades in south China between dental topography and bite performance later, all the while maintaining high levels of variability in dental complexity and convexity (Fig. 1).”

Figure 1A: The time scale in the top left of the figure looks off. Shouldn't the K-Pg be at 66 Ma (not 65 Ma) and the P-E boundary at 56 Ma (not ~54 or 55)?

We revised Fig. 1 to fix the time scale so that K-Pg is at 65.5 Ma and the P-E boundary at 56 Ma. Thank you for catching this.

Figure 1A: Is there a different y-axis scale for the variance (red line) results?

Yes, the y axes for the variance curves were missing. We added them back in. Thank you.

L628-629: As I explained above, it feels like you focused your sampling just on herbivorous/omnivorous groups, and, if true, this is an important point that should be discussed at the forefront of the paper. Does your sample truly represent the total ecological diversity of the mammalian faunas at the time?

We agree with the reviewer about the potential partial sampling of the range of ecomorphological diversity when only the most abundant clades are included in the analyses. However, we refrain from interpreting the dietary groupings represented in the dataset using an assumption of functional morphology from crown/extant clades. We added a paragraph in the introduction to bring attention to the inherent uncertainty in the ecological diversity of the dataset:

“A major challenge with expanding analyses of post K-Pg recovery to Paleocene mammal assemblages elsewhere in the world is the stratigraphically limited nature of early Cenozoic sequences that produce fossil mammals. In Asia, Paleocene localities in China represent the best studied to date 11. From the earliest Paleocene, highly regional and endemic faunas are known from a handful of sedimentary basins (Fig. S1A). Among the faunal elements, only the archaic placental clades Anagalida and Pantodonta are consistently sampled across the major subdivisions of the Paleocene 11. An additional complication with ecomorphological analysis of these early mammals is the uncertainty in their dietary ecology, as they are beyond the reach of conventional phylogenetic bracketing approaches to dietary reconstruction. Phenomic analysis of the placental radiation supports insectivory as the ancestral diet of the hypothetical placental ancestor, but uncertainty in the post K-Pg availability of insects and plants in some regions leave some doubt as to the accuracy of this ancestral state reconstruction 1. Herein we treat the archaic Paleocene taxa in our analyses as having uncharacterized diets rather than categorizing them as insectivores, herbivores, or carnivores. “

L653: Sorry if this is mentioned elsewhere, but did you avoid using teeth with especially worn or broken cusps? You might expand on how you chose teeth for your sample.

We left out this detail in the original submission. Thank you for pointing this out. We had to exclude a third of the teeth because they were too worn or broken. We added the following explanation to the methods section:

“These tooth positions were selected from a broader examination of ~300 individual teeth from 72 specimens. We vetted the specimens and excluded 99 tooth positions (~33% of teeth initially chosen for possible inclusion) from our analyses because they either (1) were partially or completely broken at the crown, (2) were in an advanced stage of attritional wear where no cusps could be identified, or (3) possessed a combination of the two aforementioned conditions.”

L654: "specimens" should be "teeth", correct? In the preceding sentence, you say that there are 200 teeth from only 48 specimens.

Corrected.

The explanation was helpful because subject headings can feel pretty confusing the first time you run into them in a library database.

Having a tips that suggest tools like a thesaurus and Visuwords was helpful because they provide practical ways to brainstorm stronger search terms. Including visual tools also supports different learning styles and approaches to research.

This section acknowledges that searching is a skill that develops over time. It makes the research process feel more realistic by showing that search strategies often need tweaking and experimentation as you go.

This chapter emphasized the importance of a good research strategy, which can be overlooked at times. Using related terms and synonyms is especially useful for broad or interdisciplinary topics.

A good reminder that research takes time and good judgment. The phrase "Be picky" really highlights the importance of being thoughtful when choosing sources, rather than just going with whatever pops up first.

This pushes back against the habit of typing a quick phrase into Google and accepting the first results. Precision searching seems especially useful for academic work where the quality and relevance of sources matters. Taking the time can help wtih quality research.

I thought this explianed why precise searching is important: better searching is not just about getting more results, but getting results that actually fit the research question.

cloture l'article avec une image familière pour parler au lecteur, avec un peu d'ironie. pas de valeur argumentative mais effet de connivence avec le lecteur.

changement de thématique. la conclusion abandonne les questions des effets sur les élèves pour ouvrir sur l'éthique de l'IA, les chatbots ...

mise en garde méthodologique

fonctionne : c'est à dire ? mesuré par quoi ? sans groupe de controle impossible d'attribuer la reussite à la robotique plutot qu'à d'autres facteurs (effet Hawthorne, motivation des enseignants ....)

encore un argument par l'exemple (un cas ne fait pas preuve ....)

il y a une confusion variable puisque l'effet attribué à la robotique pourrait etre un effet du travail en petits groupes, indépendamment de la robotique (variable confondante non controlée)

l'auteur empile et ne définis pas. catégories assez large qui normalement demande en psychologie une opérationnalisation.

assez flou, peu de précisions sur les effets de la musique via la robotique ... L'auteur n'explique pas du tout le mécanisme causal... Pas de source

La concession de la phrase précèdente est comme annulée... concéder pour mieux avancer ? Manque de précision : la plupart (combien ? )

Honnêteté de l'auteur qui reconnaît que les preuves expérimentales font défaut

ce sous titre affirme déjà un effet positif = "redonner". avant même de donner les données. C'est comme si la conclusion précèdée l'argumentation ....

Un argument par l'exemple ! un concours n'est pas une étude scientifique sur les effets de l'apprentissage ... confusion entre l'existence d'une pratique et la preuve de son efficacité

Inria propose un dispositif.... Cela prouve-t-il son efficacité pédagogique ?

le plus citée .... Par qui ? Où ? quels travaux ? encore une fois, l'article souligne un consensus sans le justifier

On remarque ici un argument épistémique pour cadrer, ce qui est une classification utile pour structurer mais il faut noter le changement de temps : empploi de deux conditionnels et puis on glisse sur un indicatif à la fin, sans le justifier ...

l'emploi du conditionnel "aurait" signale que ce cadre n'est pas établi par l'auteur, il reprendrait une classification déjà établie sans en discuter la validité.

Apparence de rigueur mais la source citée qui suit est celle de l'agence des usages des TICE qui n'est pas une revue scientifique.

On peut voir le fait de citer le catalogue de Lego comme un argument pédagogique qui relève plus de la publicité que de la recherche.

ici, un argument épistémique sur l'accessibilité. La source, Futura-Sciences, est un site de vulgarisation pour le grand public. On peut ainsi noter un écart entre les promesses du titre et la nature des sources réelles.

"vraisemblablement" est une projection non étayée, encore une fois l'affirmation est affirmée et non démontrée.

ici, on observe une progression historique qui rend la robotique scolaire inévitable par analogie (sorte de raisonnement analogique)

il s'agit d'un argument dialectique (présente cette affirmation comme un fait alors qu'il s'agit d'une opinion normative). On le voit bien avec l'utilisation du mot "doit" qui n'est pas justifiée.

présente la robotique comme un phenomene deja en cours mais qui est aussi inévitable. la question qui suit oriente vers les effets et ne questionne pas sur le dispositif en lui meme.

question plutot argumentative qui semble annoncer une approche neutre puisque c'est une revue de recherches. Une promesse épistémique que le reste du texte ne tient pas.

eLife Assessment

This important study links allelic expression imbalance with replication timing, suggesting a stochastic model for haploinsufficiency in dosage-sensitive disease. The integration of allele-specific RNA-seq and replication timing in clonal systems provides solid evidence for an association between asynchronous replication and allelic imbalance, although the scope and generality should be addressed in future work. This study will interest epigeneticists and genome regulation researchers studying replication timing and monoallelic expression, as well as developmental biologists and human geneticists concerned with clonal heterogeneity, haploinsufficiency, and variable disease penetrance.

[Editors' note: this paper was reviewed by Review Commons.]

Reviewer #2 (Public review):

Summary:

The authors pair analysis of replication timing and allele-specific expression in clonal populations of primary human cells. They combine these data with previously published data on clones from transformed human cell lines. They identify a number of genomic regions that display asynchronous replication timing in at least one clone and correlate these regions with allele-specific expression of genes within them. They also observe that several interesting gene sets, including genes that are associated with human diseases, map to asynchronously replicating regions. This is a good experimental approach that builds on already published data demonstrating the connection between allelic imbalance and replication timing.

- This is a research topic that touches on a few sub-fields of biology, and thus to make the paper more approachable we would recommend a careful edit of the text for clarity and precision of language.

- Authors point out that this is a decades-old field; we would suggest to use terminology established within the field is possible. Allelic imbalance has been referred to as AI, MAE (monoallelic expression), RMAE (random monoallelic expression) etc. The paper whose mouse data the authors make use of uses Asynchronous Stochastic Replication Timing (ASRT) instead of VERT to refer to the same phenomenon.

- Methods do not provide fully sufficient detail to fully evaluate or reproduce these experiments.

- It is helpful to show representative loci as the authors do in Fig 1F and G and Fig 2 but these panels are very densely rendered and thus difficult to process visually - even the cartoon version (1D) is thick with overlapping lines. The point that allelic imbalance is enriched in VERTs would be enhanced if the authors could present the allelic ratio for all genes found in all VERTs, demonstrating how replication timing on either chromosome affects the allelic ratio.

- The authors make the important point that VERTs are unlikely to be shared among different cell types and tissues (Fig 1i), but then find an enrichment for neuronal and immune genes in VERT regions identified in ACPs. It follows that these same genes are unlikely to be in such regions in the tissues where they are relevant. Some of the GO terms presented are too broad to suggest any biological significance to the result, even if there is statistical significance (for example, the top term for LCL clones 'Cytoplasm' is associated with 12,000 genes, and the second term for mouse clones 'Membrane' is associated with 10,000). It would be helpful to focus on GO terms lower in the GO hierarchy.

- Figure 3 highlights the association of related gene clusters with VERTs but the VERTs are assigned based on variable replication timing in just 1 or 2 clones. This is an interesting observation, but to make the point that "VERT regions frequently coincide with gene clusters in the human genome" there needs to be a systematic assessment of replication timing at all gene clusters across all clones, and a statistical test for significance.

- It is an interesting hypothesis that VERTs are conserved between species at syntenic loci. If such regions are really conserved, one would expect that replication timing at these sites would be consistently asynchronous. However the data presented shows that in human clones these VERTs can be specific to an individual donor (as in 5A) or an individual clone (as in 5H).

- The finding that VERTs coincide with neurodevelopmental disease genes in immune and cartilage cells is at odds with the previous statements and data about the tissue specificity of VERTs. In order to support the claim that neurodevelopmental disease associated genes reside in asynchronously replicating regions, and are thus more prone to allelic imbalance, it would be helpful if the authors demonstrated this phenomenon in neuronal cells.

- The authors consistently lean on sparse samples (i.e. a single clone) within a modestly sized dataset (4 clones from 2 donors each) to propose a new model for haploinsufficiency in human disease. It may well be but the consistent focus on limited elements in the data and perhaps an overreach in the interpretation makes it difficult to appreciate the very good experiments presented.

- This section refers to the revised version of the paper.

We would like to thank the authors for the changes and explanations offered. Although we don't fully agree with a few answers offered, overall the answers and changes in the manuscript have significantly improved the work presented. As such it should be of interest to many readers.

Author response:

The following is the authors’ response to the original reviews

General Statements

We thank the reviewers for their thoughtful and constructive comments, which substantially improved our manuscript. In response, we have revised the text and figures throughout to address the points raised. Specifically, we have:

i. Refined our definition of Inactivation/Stability Centers (I/SCs): We limit this designation to loci where both Allelic Expression Imbalance (AEI) and Variable Epigenetic Replication Timing (VERT) were detected, either in the present study or in previously published work.

ii. Expanded methodological clarity: We provide detailed descriptions of how VERT regions were identified, annotated, and quantified, including thresholds for allelic imbalance, replication timing variability, and sampling depth. We also justify the ≥80% AEI cutoff, which is based on recently published studies showing that modest allelic biases can have biological and clinical significance.

iii. Enhanced benchmarking and validation: In addition to the analysis of X inactivation in female ACP cells, we now include comparisons between imprinted and non-imprinted regions to benchmark the magnitude of allelic replication timing imbalance, demonstrating that the magnitude of imbalance observed at non-imprinted VERT regions is comparable to known imprinted regions.

iv. Address tissue specificity and sampling limitations: We now discuss how the data derived from a limited number of clones, tissues, and individuals support the identification of robust AEI and VERT patterns.  In the future, additional tissues and individuals will be required to capture the full diversity of I/SC regulation.

v. Clarify biological relevance: We have expanded our discussion to highlight the consistency of AEI findings across cell types, including examples of genes implicated in neurodevelopmental and neurodegenerative disorders, and we clarify our model of how I/SC regulation contributes to haploinsufficiency, variable expressivity, and incomplete penetrance in human disease.

vi. Improved figures and supplemental data: We have updated figure legends for clarity, added a new supplementary figure benchmarking imprinted regions, added supplementary tables containing: the full description of our GO analysis, the list of I/SCs where we have detected both VERT and AEI, the ratios of the number of transcripts derived from early and late replicating alleles for the I/SCs illustrated in all figures, and we have cross-referenced all supplementary tables.

Point-by-point description of the revisions

Reviewer 1:

The existence of VERT regions is well supported, but the number of regions called as ISCs may be inflated by permissive thresholds (e.g., AEI {greater than or equal to} 0.8 or {less than or equal to} 0.2 in a single clone). This risks conflating transient stochastic differences with stable ISCs.

We selected the >80% (or <20%) allelic imbalance threshold, along with the requirement of at least one biallelic clone, as our criterion for significant AEI. This choice was guided by a recent study demonstrating that allelic imbalance, as low as a 65%/35%, is enough to effect disease penetrance in humans (Nature 2025; 637:1186–1197). For completeness, results obtained using more stringent thresholds (>90% and >95% imbalance) are presented in Supplementary Table 2.

Furthermore, it is unlikely that transient stochastic differences in allelic expression, such as those detected by single-cell RNA sequencing assays (Nat. Rev. Genet. 2015; 16:653–664), would be captured by our approach. Each clone in our study was expanded from a single cell to over one million cells before both RNA-seq and Repli-seq analyses, effectively averaging out transient transcriptional and/or replication fluctuations, and thus reflecting stable, mitotically heritable epigenetic states.

Reviewer 1:

More robust approaches would include using magnitude of imbalance, annotating VERTs by genomic location, applying stricter thresholds for replication timing, and benchmarking AEI distributions against the X chromosome.

All VERT regions identified in this study were annotated according to both the magnitude of allelic imbalance and their genomic coordinates, using 250 kb windows for the human samples and 50 kb windows for the mouse samples (see Supplementary Tables 1 and 6). Figure 1c directly compares the magnitude of imbalance, defined as outliers in the standard deviation, for both allelic replication timing and allelic expression across autosomal and X-linked loci in female ACP cells.

In addition, we detected allelic replication asynchrony at 12 known imprinted loci, and the standard deviation of replication timing at these loci, measured in 250 kb windows, is comparable to that observed across the >350 VERT regions detected at non-imprinted sites. For comparisons, we have highlighted the imprinted regions with + symbols in Figures 1e, 2d, 3c, 6g, 7e, 7g, and we have highlighted the imprinted regions in Supplemental Table 1, and in the Data Source files. For additional comparisons, we have included Supplemental Figure 1 to illustrate the magnitude of replication timing imbalance and allele-specific gene expression at two autosomal imprinted regions.

Reviewer 1:

Figures and text would benefit from improved clarity: axis labels are missing in places (e.g., Fig. 1c, Fig. 2g), legends should explain chromosome arm colors, and cluttered figures such as Fig. 1j could be re-visualized for interpretability.

Figure labels have been added to Figs. 1c and 2g, and legends modified for clarity.

Reviewer 1:

“…the claim of cell-type specificity is not convincingly demonstrated given the small sample size (n=4) and strong batch confounding between lymphoblastoid and cartilage progenitors.” And “Hierarchical clustering is confounded by batch and based on presence/absence calls that lack quantitative resolution.”

We agree that the limited number of individuals and clones, as well as the comparison between only two distinct tissue types (LCLs and ACPs), have quantitative limitations. Our primary intent was to evaluate whether any I/SCs were shared between independently derived clonal cell lines from different tissues to determine whether there is evidence of tissue-specific I/SC usage, rather than to make quantitative claims about global cell-type specificity.

To address this concern, we have replaced the hierarchical clustering analysis, in Figure 1i, with a Venn diagram that more directly illustrates the overlap and tissue-specific distribution of VERT regions detected in the different clonal sets. This revised representation avoids assumptions about clustering relationships and removes batch-driven bias, while still conveying the key observation that many VERT regions are shared across tissues and others appear tissue-restricted.

Reviewer 1:

While syntenic VERT regions across mouse and human are intriguing, they complicate interpretation of strong clustering by cell type. Sampling depth may also have exaggerated allelic imbalance calls.

We note that the human LCLs used in our study are B cells, and immunoglobulin gene rearrangements were used to confirm the clonal uniqueness of each line. Similarly, the mouse replication timing data analyzed here was generated from pre-B cells, which also undergo immunoglobulin gene rearrangements. Thus, both the human LCL and mouse pre-B cell datasets were derived from B-cell lineages, providing a consistent cellular context for comparative analysis.

Sequencing depth is an important consideration for all variant base calls. Without fully haplotype-resolved genomes, previous studies relied on calculating per-SNP calls of allelic imbalance based on reads covering a single nucleotide locus. To improve sequencing depth supporting the identification of VERT and AEI regions, we utilized haplotype-resolved genomes that allowed all informative allele-specific reads to be pooled across all heterozygous SNPs within genomic windows or expressed genes. For AEI, we set a minimum threshold of 20 informative allele-specific reads per gene, a minimum FDR-corrected p-value of <=0.05, and a minimum of 80% vs 20% allelic imbalance. Importantly, a recent study showed that allelic imbalance as low as a 65%/35% is clinically relevant in humans (Nature 2025; 637:1186–1197). We reiterate that more stringent thresholds (>90% and >95% imbalance) are presented in Supplementary Table 2.

Reviewer 1:

Gene set enrichment analysis should be restricted to avoid inflated significance from overly broad categories.

Reviewer 2:

Some of the GO terms presented are too broad to suggest any biological significance to the result, even if there is statistical significance (for example, the top term for LCL clones 'Cytoplasm' is associated with 12,000 genes, and the second term for mouse clones 'Membrane' is associated with 10,000). It would be helpful to focus on GO terms lower in the GO hierarchy.

We now include our complete Gene Ontology analysis, with more specific biological categories, in Supplemental Table 5.

Reviewer 2:

Allelic imbalance has been referred to as AI, MAE (monoallelic expression), RMAE (random monoallelic expression) etc. The paper whose mouse data the authors make use of uses Asynchronous Stochastic Replication Timing (ASRT) instead of VERT to refer to the same phenomenon. Creating unnecessary jargon makes the paper more difficult to read and adds needless complexity to an already complex field.

While we agree that allelic expression imbalance has been described by different investigators using many different phrases, we believe that MAE, RMAE and AI do not represent an accurate description of the phenomenon. In our study [and our previous study; Nat Commun. 2022; 13(1):6301] we used clonal analysis of allele-specific expression and found that while some clones display equivalent levels of expression between alleles of a given gene (i.e. bi-allelic expression) other clones express only one allele (i.e. mono-allelic expression), and yet other clones have undetectable expression (i.e. silent on both alleles). This pattern of allele-restricted expression indicates that each allele independently adopts either an expressed or silent state. Importantly, because these expression states are mitotically stable, allele-autonomous, and independent of parental origin, we refer to the choice of the expressed allele as stochastic. Given this variability, we believe that the phrase “Allelic Expression Imbalance” (AEI) represents a more accurate descriptor for this phenomenon. We also point out that “Allelic Expression Imbalance” has also been used by other investigators >120 times in the Pubmed database.

In addition, the replication asynchrony that exists at these loci is not consistent with purely ASynchronous Replication Timing (ASRT) between alleles. We found that each allele can independently adopt either earlier or later replication timing in different clones. This variability results in some clones exhibiting pronounced asynchrony between alleles, while in others, the two alleles replicate synchronously, with both adopting either the earlier or later timing state. As reported in our previous study (Nat. Commun. 2022; 13:6301), this behavior reflects a stochastic and allele-autonomous process, leading us to describe these loci as exhibiting Variable Epigenetic Replication Timing (VERT), which we believe is a more accurate descriptor of this phenomenon.

Reviewer 2:

The point that allelic imbalance is enriched in VERTs would be enhanced if the authors could present the allelic ratio for all genes found in all VERTs, demonstrating how replication timing on either chromosome affects the allelic ratio.

The stochastic nature of allelic expression and replication timing observed at VERT loci indicates that each allele independently acquires its epigenetic state. In addition, there are typically more than one transcription unit, both protein coding and non-coding, within each VERT region, and each transcription unit also acquires its expressed or silent state independently.  Therefore, the expressed or silent status of one allele of a transcription unit does not predict the replication timing or expression status of the same or opposite allele of any other transcription unit within the VERT region. Accordingly, the Early/Late pattern of replication timing that we detect, both in this study and in our previous work (Nat. Commun. 2022; 13:6301), is not correlated with which allele is transcriptionally active. This supports our conclusion that asynchronous replication timing is not a downstream consequence of monoallelic transcription, but rather an independent epigenetic feature of I/SCs. Regardless, because each transcription unit is independent, we provide the expression ratios for all transcripts that are generated from the VERT regions for the coding and non-coding transcription units in Figures 1, 2, and 6; shown in Supplemental Table 9. This analysis indicated that 4,017 informative reads were derived from the earlier replication allele and 3,161 informative reads were derived from the later replication allele, generating an allelic ratio of 1.3 (early/late) and a binomial P value of 1.0.

In addition, a similar analysis of imprinted loci reveals that even at genomic regions with parent-of-origin–specific expression, the replication timing of each allele does not align with transcriptional activity, i.e. both early- and late-replicating alleles can be transcriptionally active, depending on the gene. This observation is consistent with the complex organization of many imprinted domains, where genes on opposite alleles exhibit reciprocal expression patterns. To illustrate this point, we now include Supplemental Figure 1 demonstrating that imprinted loci harbor genes expressed from both the earlier- and later-replicating alleles. In addition, quantification of the total number of transcripts at the DLK1/MEG8 imprinted locus (Supplementary Figure 1a-1c) indicates that the ratio of transcripts derived from the early versus late replicating alleles is equivalent (i.e. a ratio of 1.0; See Supplemental Table 9).

Reviewer 2:

Figure 3 highlights the association of related gene clusters with VERTs but the VERTs are assigned based on variable replication timing in just 1 or 2 clones. This is an interesting observation, but to make the point that "VERT regions frequently coincide with gene clusters in the human genome" there needs to be a systematic assessment of replication timing at all gene clusters across all clones, and a statistical test for significance.

Our intent in Figure 3 was not to suggest that all gene clusters are subject to VERT and AEI, but rather to highlight that several well-characterized multigene families that are known to exhibit AEI, such as olfactory receptor, protocadherin, and HLA gene clusters, coincide with VERT regions at their genomic locations. These examples serve as representative illustrations demonstrating that I/SC-associated regulation occurs at established AEI loci organized in gene clusters.

To clarify this point, we have revised the text to explicitly state that Figure 3 presents illustrative examples of known AEI-associated gene clusters overlapping with VERT regions, rather than a comprehensive or statistically exhaustive analysis of all gene clusters across the genome.

Reviewer 2:

It is an interesting hypothesis that VERTs are conserved between species at synentic loci. If such regions are really conserved, one would expect that replication timing at these sites would be consistently asynchronous. However the data presented shows that in human clones these VERTs can be specific to an individual donor (as in 5A) or an individual clone (as in 5H).

As discussed in our Limitations Section, our analysis was restricted to a limited number of cell types, clones, and individuals, which may not capture the full diversity of I/SC usage across tissues and populations. While our dataset was sufficient to identify robust patterns of AEI and VERT, it likely represents only a subset of the broader landscape of I/SC regulation in both humans and mice. We anticipate that future studies incorporating a wider range of tissues, individuals, and clonal analyses will uncover an even greater degree of conservation and diversity in I/SC usage across genomes.

Reviewer 2:

In order to support the claim that neurodevelopmental disease associated genes reside in asynchronously replicating regions, and are thus more prone to allelic imbalance, the authors would need to demonstrate this phenomenon in neuronal cells.

We make two points that address this critique: First, many of the neurodevelopmental disease genes located within or adjacent to VERT regions are not exclusively expressed in neuronal cells and have previously been shown to exhibit AEI in non-neuronal contexts. For example, Gimelbrant and Chess (Science, 2007; 318:1136–1140) demonstrated AEI of the Parkinson disease genes SNCA and LRRK2 in lymphoblastoid cell lines (LCLs), and in our previous study, we detected AEI of DNAJC6, another Parkinson disease gene, also in LCL cells (Nat. Commun. 2022; 13:6301). In the present study, using cartilage progenitor cells, we identified VERT and AEI of several epilepsy-associated genes, including SCN1A, SCN2A (Fig. 6b), GABRA1(Fig. 6e), and SAMD12 (Fig. 6j), as well as a gene implicated in autism and neurodevelopmental disorders, SEMA5A (Fig. 5c), indicating that these genes are not exclusive to neuronal cell types.

Second, independent studies from the Dr. E. Heard laboratory have provided further evidence that AEI occurs in neuronal lineages. Using mouse neural progenitor cells (NPCs), they identified genes subject to AEI (Dev. Cell, 2014; 28:366–380) and they later evaluated AEI of syntenic human neurodevelopmental disease genes, including Snca, App, Eya4, and Grik2 (Nat. Commun. 2021; 12:5330). In addition, and consistent with our use of AEI, they used the phrase “Allelic Expression Imbalance” to describe the epigenetic expression biases at these genes.

Together, these findings reinforce that AEI, and by extension I/SC regulation, is not restricted to specific cell types, but rather represents a generalizable mechanism of stochastic epigenetic regulation that includes genes relevant to neurodevelopment and disease.

Reviewer 2:

However, the authors consistently lean on thin evidence (i.e. a single clone) within a modestly sized dataset (4 clones from 2 donors each) to propose a new model for haploinsufficiency in human disease. The consistent focus on limited elements in the data and perhaps an overreach in the interpretation makes it difficult to appreciate what is in fact a very good experiment.

We agree that our analysis was conducted on a modest number of clones and individuals, which we explicitly acknowledge as a limitation of the present study. However, several key points support the robustness and broader relevance of our conclusions:

i. Clonal Design and Replication: The strength of our approach lies in its clonal resolution. Each clone represents a single-cell–derived population expanded to over a million cells, enabling direct detection of stable, mitotically heritable allele-specific epigenetic states that would not be apparent in population-averaged data. Importantly, many of the VERT regions we identified are shared between independent clones from different donors and across distinct cell types (ACP and LCL), demonstrating reproducibility and biological consistency.

ii. Cross-Species Validation: We further identified syntenic VERT regions in mouse pre-B cell clones, including at loci known to exhibit AEI in prior studies, providing independent validation and evolutionary conservation of the phenomenon.

iii. Integration with Published Evidence: Our findings extend prior observations of AEI and VERT (e.g. Gimelbrant et al. Science 2007; Heskett et al. Nat. Commun. 2022) and are fully consistent with known stochastic allelic expression imbalance of autosomal genes. We also draw parallels with the absence of cellular selection mechanisms that dictate dominant inheritance patterns for loss of function alleles for X linked disease genes (reviewed in: J Clin Invest, 2008, 20-23; and Nat Rev Genet. 2025, 26, 571–580). Our proposed model linking I/SC regulation to haploinsufficiency is therefore a synthesis of our results with an extensive body of published data, not an inference drawn from isolated observations.

iv. Scope and Framing: We have revised the manuscript to clarify that our proposed model represents a mechanistic framework, not a definitive or exclusive explanation, for how stochastic allelic regulation could contribute to dosage-sensitive disease phenotypes. We also explicitly discuss the need for larger datasets and additional tissues to refine and test this model.

In summary, while we recognize the limited sampling depth inherent to clonal analyses, the consistency of our observations across donors, cell types, and species, together with prior corroborating studies, supports the validity of the conclusions and justifies the broader conceptual implications.

Great comment! It's essential to become familiar with using the filter options for effectively finding research material.

This was helpful; including both keyword and subject searching explanations helps readers understand why some searches yield better results than others.

Each university has numerous resources, and if they are not utilized—especially in the areas of liberal arts and libraries—they may be at risk of review to determine their necessity. Understanding which resources are available can be helpful in ensuring they are used effectively.

Understanding this distinction is beneficial because library catalogs and databases can often be confused with one another. This section clearly explains why searching for a journal title in a catalog differs from locating specific articles.

This seems like a hot topic and a good conversation point to foster discussion of material.

This section effectively illustrates the advantages of using databases for academic research compared to general web searches. Providing an example that contrasts a topic searched on Google with one searched in a scholarly database would further clarify and emphasize these differences.

This stood out to me because the Venn diagram and the comparison made the differences between these tools much easier to grasp. It’s easy to confuse them—I did too—but the readings really helped clarify how they differ.

This definition cleared things up for me because I’ve always found databases a bit confusing. They come up a lot in research assignments, but nobody really explains how they work. I also like how it compares them to print indexes and bibliographies; it really ties together the old school and new school ways of doing research.

The “MENU” link is short and easy to understand, which supports clear navigation. This is helpful for users who want to quickly find food and drink information without searching through the whole website.

The “CAREERS” link is clearly labeled, which is good because users can understand its purpose before clicking. However, the small font size in the navigation bar may be harder to read for some users with low vision.

This heading is large and visually clear, which makes the main purpose of the page easy to understand. The strong contrast between the white text and dark background also improves readability.

The “SET YOUR LOCATION” option is useful because it helps users quickly find nearby restaurant information. However, it may be difficult for some users if the dropdown is not easy to navigate by keyboard or screen reader.

The “LOCATIONS” link is clearly labeled, which is good for accessibility because users can understand where the link will take them. However, the text is quite small in the navigation bar, so increasing the font size could make it easier for users with low vision to read.

This is a great reminder that research help isn’t just about search tools. There are so many resources in the library that people often miss, and they hold a ton of knowledge about the materials you’re looking into. Plus, getting a fresh perspective can really boost your research.

Providing support tailored to specific assignments is incredibly beneficial for students who feel overwhelmed at the start of a research project. Including a screenshot of a course-specific LibGuide would enhance this section, making it more practical and easier to visualize.

This is interesting because a lot of people think that academic library resources are only for students who are currently enrolled. The explanation shows just how useful these guides can be for anyone looking to learn on their own or find help with research.

The description does a great job of showing how libraries offer a lot more than just regular books and databases. By thinking of LibGuides as little mini-websites, it makes it easier to understand what they do.

eLife Assessment

This important study highlights how cell size influences various cellular responses, with a particular focus on ferroptosis. The evidence presented is convincing, employing multiple model systems and experimental approaches to support the conclusions. This work will be of significant interest to the fields of cell size, ferroptosis, and cancer biology.

[Editors' note: this paper was reviewed by Review Commons.]

Reviewer #1 (Public review):

Summary:

The study by Zatulovskiy et al. examined how cell size influences cell susceptibility to ferroptosis. The authors found a size dependence specifically for ferroptosis-inducing drug Era2, but not for other drugs. Using various human cell lines (HMEC, HT 1080, RPE 1), the authors generated populations of small and large G1 cells by FACS, CDK4/6 inhibition (palbociclib), or inducible cyclin D1 knockdown, and measured cell susceptibility to ferroptosis. Larger cells were more resistant than smaller cells. Mechanistically, larger cells showed reduced plasma membrane lipid peroxidation, higher glutathione concentrations, and changes in relevant cellular proteins levels, as analyzed using previously published data. Deleting ACSL4, which is involved in ferroptosis, partly eliminated the size dependence of ferroptosis. The work concludes that cell size is a key determinant of ferroptosis susceptibility. Overall, this work expands our understanding of how cell size is correlated with functional properties of cells, which can have implications for biomedical sciences.

Strengths:

The study establishes a credible link between cell size and susceptibility to ferroptosis, as induced by Era2. Experimental replication is sufficient, and key conclusions rely on data from multiple cell lines and on multiple approaches to manipulate cell size. This suggests that the conceptual findings made in this paper could reflect a more fundamental feature of mammalian cells. In addition, this work provides an interesting contrast to another recent study about size-dependency of ferroptosis (https://doi.org/10.1016/j.isci.2025.112363), where increased cell size heightened sensitivity to the GPX4 inhibitor RSL3.

Original Weaknesses:

Disentangling cell size effects from other confounding factors, such as the cell cycle or overall metabolic rate, is challenging, and the authors have managed to qualitatively prove that cell size influences Era2-induced ferroptosis. However, the quantitative nature of this link between cell size and susceptibility to ferroptosis remains somewhat unclear due to the confounding factors that are present in many of their experiments. Notably, the quantitative nature of this link could also be cell type and growth condition -dependent, which remain to be investigated in detail. It should also be noted that this work focused on cell culture studies, and it remains unclear how much the findings of this paper could influence therapeutic strategies in vivo.

Comments on revised version:

I would first like to emphasize that I find this work solid, and I think the authors have done good work with the revisions.

My only remaining recommendation is that the authors aim to more carefully examine the magnitude of the observed cell size-dependency in ferroptosis susceptibility. Their manuscript contains several experiments where the quantitative nature of this link remains unclear due to confounding factors, such as the cell cycle. For example, in Fig 2B&C, it seems that accumulation of cells in G1 (from ~60% to ~95%) decreases ferroptosis equally to the effect caused by cell volume doubling (from day 2 to day 4 of palbo treatment), suggesting that cell cycle has a much more pronounced effect on ferroptosis than cell size (especially when considering the size change from day 0 to day 2). However, the magnitude of the cell size effect is not consistent between all experiments shown. This is not surprising, as the authors use different approaches to changing cell size and different cell lines, but it makes the work more qualitative than quantitative. Notably, another confounding factor is the cell's metabolic/biosynthetic rate. It seems reasonable to assume that prolonged palbociclib treatment will decrease metabolic and protein synthesis rates (normalized to cell size), and this could make the cells less susceptible to ferroptosis. The rapamycin treatment results shown by the authors also support this notion. One approach to examining this could be to grow cells in various growth conditions to manipulate their growth & metabolic rate.

Reviewer #2 (Public review):

Summary:

The authors set out to understand how cell phenotypes differ depending on the size of the cell, specifically here how cell size affects cell death. Using human cell lines (HMEC, HT-1080, RPE-1), the authors examined cell size through FACS sorting, CDK4/6 inhibition and inducible cyclin D1 knockdown. They identify that larger cells are more resistant to ferroptosis induced by system xc^- inhibition (erastin2), but more sensitive to GPX4 inhibition (RSL3), highlighting pathway-specific size dependencies.

Mechanistically, larger cells exhibited:

- Higher glutathione levels, supporting lipid peroxide detoxification

- Increased ferritin expression, promoting iron sequestration

- Lower ACSL4 levels, reducing incorporation of peroxidation-prone lipids

The findings are supported by high-throughput microscopy, flow cytometry (BODIPY-C11 lipid peroxidation assays), and proteomic analyses. The study concludes that cell size influences proteome composition and metabolic capacity, thereby shaping cell death decisions, an insight with implications for aging, cancer, and ferroptosis-based therapies.

Major Strengths:

- use of multiple cell lines to validate their findings

- use of multiple, complimentary approaches

- well designed screen and experiments throughout

- clearly written, logical flow and easy to follow

- relevance for multiple fields

Weaknesses:

- Lack of in-depth mechanistic investigation

- Experiments are all in vitro and so, as yet, it is uncertain what the in vivo consequence would be

General Assessment:

This study presents a mechanistic link between cell size and ferroptosis susceptibility. Using high-throughput microscopy, proteomics, and genetic perturbations across multiple human cell lines, the authors demonstrate that larger cells are more resistant to ferroptosis induced by system xc^- inhibition (erastin2). This resistance is attributed to elevated glutathione production, increased ferritin-mediated iron sequestration, and reduced ACSL4-dependent lipid peroxidation. The experimental design is rigorous and multifaceted, with consistent results across cell types and size manipulation methods. While the study is limited to in vitro systems, its conceptual and mechanistic insights lay the groundwork for future in vivo and translational investigations.

Advance:

This work is the first to systematically show that cell size directly influences ferroptosis susceptibility via proteome scaling. It reconciles previous findings that large cells are sensitized to GPX4 inhibition (RSL3) by demonstrating that the ferroptosis pathway targeted system xc^- vs GPX4 determines the direction of size-dependent vulnerability. The study provides a conceptual advance by positioning cell size as a regulatory axis in cell death decisions, and a mechanistic advance by identifying size-dependent changes in glutathione metabolism, ferritin levels, and ACSL4 expression.

Audience:

This research will be of interest to specialists in cell death, ferroptosis, redox biology, and cancer biology. It also holds relevance for aging researchers and translational scientists exploring ferroptosis-based therapies. The findings may influence how cell size heterogeneity is considered in therapeutic design, particularly in oncology and senescence-targeting strategies.

Comments on revised version:

We have no additional comments after revision. Thank you for addressing our initial queries.

Reviewer #3 (Public review):

In this manuscript, Zatulovskiy and colleagues elaborate on their previous work describing cell size-dependent changes in the proteome by investigating whether these changes can be correlated in differences in cell physiology. Using a cleverly-designed high throughput screen, they searched for compounds that differently-sized cells display differential sensitivity towards. Their primary hit, Era2, is involved in the ferroptosis pathway and serves as the starting point for a detailed study of how excess cell size protects cells from ferroptosis-induced cell death via: 1) lower concentrations of ACSL4 (which produces peroxidation-prone PUFAs), 2) increased ferritin concentrations, and 3) increased GSH concentrations.

Overall, the experiments in this manuscript are well-designed and interpreted. It is an extremely well-written manuscript with a clear trajectory of logic.

Comments on the revised version:

The authors have addressed my original concerns adequately. I do not need to see it again, if there are further revisions.

Author response:

General Statements

We were pleased to see that all three reviewers support publication after revision. No one questions the premise that cell size influences ferroptosis susceptibility. The main concerns fall into two categories: (A) disentangling “Cell size vs cell cycle”, which is the biggest issue for Reviewer #1 and partially for #3. (B) Additional mechanistic tests including SLC7A11 and ferritin functional tests (Reviewer #2) and lysosomal iron (via LysoRhoNox) and some further ACSL4 experiments (Reviewer #3). Other reviewer concerns are more minor.

In our revision, we have addressed the reviewer’s specific criticisms with additional experiments as described below. We believe the constructive feedback from peer reviews helped us to significantly extend our mechanistic findings and strengthen the manuscript through revision.

Point-by-point description of the revisions

Reviewer #1:

Summary:

The study by Zatulovskiy et al. examined how cell size influences cell susceptibility to ferroptosis. The authors found a size dependence specifically for ferroptosis-inducing drug Era2, but not for other drugs. Using various human cell lines (HMEC, HT 1080, RPE 1), the authors generated populations of small and large G1 cells by FACS, CDK4/6 inhibition (palbociclib), or inducible cyclin D1 knockdown, and measured cell susceptibility to ferroptosis. Larger cells were more resistant than smaller cells. Mechanistically, larger cells showed reduced plasma membrane lipid peroxidation, higher glutathione concentrations, and changes in relevant cellular proteins levels, as analyzed using previously published data. Deleting ACSL4, which is involved in ferroptosis, partly eliminated the size dependence of ferroptosis. The work concludes that cell size is a key determinant of ferroptosis susceptibility.

My major concerns about this work focus on whether many of the results reflect cell size or cell cycle effects, and whether the FACS-based size-scaling analyses have some misleading features to their design & presentation. If these concerns can be addressed with new experiments, then the conclusions of this paper are justified. If these concerns cannot be addressed, then the authors should more directly acknowledge the alternative hypothesis that cell cycle effects may explain many of their results.

The experiments seem to be replicated sufficiently, and most conclusions rely on data from multiple cell lines. My minor comments focus on needs to provide statistics and method details, and on suggestions on how to improve text clarity, but these edits are easily done and don't require new experiments. Overall, this is an interesting study, and it should be published once the concerns below are addressed.

Major comments:

In experiments reported in Fig 1 and 2A, the authors sort small and large cells in G1, plate them, and later start the drug treatments & cell monitoring. Are these cells actively cycling (progressing in the cell cycle), and how fast? The large cells are likely to enter S phase earlier than the small cells, so by the time that the authors start their drug treatments, they may be comparing cells in different cell cycle stages, which could influence drug sensitivity more than cell size (as the authors also suggest later in Fig 2). This needs to be controlled for.

Furthermore, even if the cells remain in G1 after sorting until the drug treatments are started, the authors should address the fact that the drugs are present for a long time, thus targeting the cells in various cell cycle stages.

We agree with the reviewer that the cell cycle stage could affect ferroptosis susceptibility and could be a confounding effect in asynchronous cells. One of us (Dixon) reported the cell cycle effects on ferroptosis previously, and we observe them in this manuscript too (Fig. 2B,C,E). We now state this more clearly both in the Results and in the Discussion sections, where we write:

Line 159: “We note that non-arrested cells had a lower susceptibility to Era2-induced ferroptosis compared to cells that were arrested in G1 for 2-3 days, despite being smaller in size. This is likely due to the difference in the fraction of cells in different cell cycle phases between arrested and non-arrested conditions since cells in S/G2/M phases are known to be more resistant to ferroptosis than cells in G0/G1 phases (Rodencal et al, 2024; Kuganesan et al, 2023)”

Line 533: “Cells in G1 phase of the cell cycle were reported to be more susceptible to ferroptosis (Rodencal et al, 2024; Kuganesan et al, 2023), which suggested that ferroptosis inducers could be used in combination with cancer drugs, like the CDK4/6 inhibitor palbociclib, that arrest cells in G1 phase of the cell cycle (Herrera-Abreu et al, 2024). However, while CDK4/6 inhibitors arrest cells in G1, they do not inhibit cell growth, such that the longer they are arrested, the larger the cells grow (Lanz et al, 2022; Crozier et al, 2023; Manohar et al, 2023). This results in a complex, nonmonotonic ferroptotic response dynamics in cells treated with CDK4/6 inhibitors (Fig. 2B,E). Just following CDK4/6 inhibitor treatment, as more and more cells are arrested in G1 phase, cells become more sensitive to both RSL3- and erastin-induced ferroptosis (Kuganesan et al, 2023; Rodencal et al, 2024). However, the longer the cells are arrested, the larger they become, which further promotes their susceptibility to RSL3 (Fig. S1B) but reduces their susceptibility to Era2-induced ferroptosis (Fig. 2B). The fact that the cell cycle arrest and cell size increase have opposing effects on Era2-induced ferroptosis susceptibility could explain why different studies reported seemingly contradictory results, where sometimes an increased and sometimes a decreased or unchanged sensitivity to system x_c^- inhibitors was observed depending on the cell type, duration and type of cell cycle arrest (Lee et al, 2024; Kuganesan et al, 2023; Rodencal et al, 2024). Such complex interplay between the cell cycle and cell size effects on ferroptosis suggests that combination therapies utilizing CDK4/6 inhibitors and ferroptosis inducers would have to carefully choose a dosage schedule.”

Given the potentially confounding effects of the cell cycle in cycling cells sorted by size, we performed an additional experiment, in which RPE-1 cells were pre-treated with the CDK4/6 inhibitor palbociclib to synchronize them in G1 phase prior to treatment. These cells were then continuously exposed to palbociclib during the Era2 treatment (Fig. 2C-E). RPE-1 cells pretreated with palbociclib for 2 and 4 days had the same cell cycle distribution with 94% of cells being arrested in G1, but with different sizes. Cells treated with palbociclib for 4 days were significantly larger and more resistant to Era2.

Additionally, in the experiment shown in Fig. 5E,F, where we FACS-sorted WT and ACSL4 KO HMEC cells by cell size, and then measured Era2 susceptibility, we pre-treated the cells with palbociclib for 24 h to synchronize them in G1 prior to the sorting. We then cultured the cells in the presence of palbociclib during the Era2 treatment to avoid the cell cycle effects observed in Fig. 2. In this case, we still observe that larger cells are more resistant to Era2, consistent with our conclusion that cell size protects against Era2-induced ferroptosis.

Can the G1 arrest-driven changes in drug susceptibility (Fig 2 C-D) be attributed to cell size? Can the authors rescue the palbociclib treatment with rapamycin or other growth inhibitors that allow size to remain small during G1 arrest?

We have attempted to perform these experiments, but when we co-treated the cells with palbociclib and mTORC inhibitors, but observed variable results, which are likely due to the fact that prolonged mTORC inhibition itself rewires cellular metabolism and reduces cell susceptibility to ferroptosis, as one of us (Dixon) found previously (Armenta et al. (2022), Ferroptosis inhibition by lysosome-dependent catabolism of extracellular protein. Cell Chemical Biology 29: 1588-1600.e7). Our results were consistent with this previous report and is now included in a new supporting figure panel (Fig. S3C):

Thus, upon palbociclib+rapamycin co-treatment there seems to be a competition between cellsize-mediated and metabolism-mediated effects of mTORC inhibition on ferroptosis, which leads to variable outcomes.

In Fig 2E-F, is the cell cycle distribution of the samples influenced by CCND1 shRNA induction? Are the drug sensitivity effects due to cell size or cell cycle changes?

The CCND1 manipulation model is extensively characterized in our recent work cited in this manuscript (You et al. (2025), Cell size-dependent mRNA transcription drives proteome remodeling. 2025.10.30.685141 doi:10.1101/2025.10.30.685141). Indeed, CCND1 shRNA cells have a slightly elongated G1 phase due to a ~30% reduction in Cyclin D1 concentration: the G1 fraction changes from ~70% in wild-type to ~80% in CCND1 shRNA cells, which could potentially affect the ferroptosis susceptibility, but the additional results obtained on synchronized RPE-1 cells, described above (Fig. 2C-E), support the conclusion that the primary effect on Era2 sensitivity is due to cell size.

Can the authors address the meaningfulness of the FACS-based size-scaling results in cases where cell-to-cell variability is very large? For example, in Fig 4D&G, the results are so variable even in identically sized cells that the importance of the size-scaling pattern seems questionable.

We do observe variability in fluorescent probe-based measurements of GSH and lipid oxidation, which could be due to biological (natural cell heterogeneity) and/or technical (low sensitivity of the probes) reasons. However, when we look at binned data and compare the mean values ± s.e.m. for each bin, we observe a robust and reproducible trend (black line with dark-grey shaded area), even though the SD is quite broad (lighter shaded area). We believe such trends are meaningful when describing cell death in probabilistic terms as we do. I.e., the GSH measurement might not be precise enough to predict cell death for a given individual cell, but the statistical trend is clear and these measurements help predict cell death probabilities for cells of different sizes.

In Figs 4B-D, the cell size axis seems to have over 4-fold size variability, but when the authors show the analysis of this data (Figs 4E-G) the variability is only 2-fold. What was excluded and on what basis?

To address this point, we have now clarified in the Methods section how the data were processed and what data points we excluded from this analysis:

Line 671: “For all binned flow cytometry data plots, the cells below the 2nd and above the 98th cell size percentiles were excluded to remove the extreme outliers. Then, the remaining data were binned by size and plotted as background-corrected average fluorescence intensity for each bin against the bin’s average cell size. Bins with fewer than 200 cells were excluded from the analysis to reduce noise.”

Typically, such pre-processing reduces the size range, mostly from the large-cell end, because of the long right tail of the size distribution containing a few very large cells.

Based on the methods section & figure legends of Fig 4B-I, the RPE cells were not pre-sorted to include only G1 cells, nor did the assay account for cell cycle differences. How can these data be used to explain results from earlier figures, where analyses were exclusively focused on size differences in G1?

This is a valid point: Cells in the GSH measurement experiment were not gated by Hoechst signal for G1 phase because the channel normally used for Hoechst staining was in this case occupied by the MCB probe. However, given the data in Fig. 4A,B showing that the GSH production machinery is superscaling when measured specifically in G1-phase cells, we believe the flow cytometry data in Fig. 4C-J showing GSH concentration increasing with cell size across the whole cell cycle is very likely true for G1 cells as well.

Minor comments:

I recommend clarifying in the early introduction that all size changes discussed are in the absence of DNA content increase.”

We have now clarified this in the introduction (Line 41 and Line 81).

The introduction seems to cite primary research and review paper in the same sentences, which is a bit misleading as the reviews don't seem to add new evidence.

We have removed review citations where they did not provide additional context.

OPTIONAL

In the second introduction paragraph, consider the classification/description of the three different mechanisms. Currently, it seems that these mechanisms are not independent of each other, and the details provided about each mechanism are inconsistent.”

We have now modified this paragraph to make the description more consistent.

Please provide statistics for the IC50 values reported based on Fig 1C. Were small and large cells statistically different? Are the IC50 values reported as +/- standard deviation or some other metric?

This has now been clarified in the text as follows:

“For example, at the 72 h time point, the Era2 IC50 was 28 ± 11 µM (mean ± SD) for large cells versus 2.0 ± 1.4 µM for small cells (Student’s t-test: p = 0.039) (Fig. 1C).”

Providing more insight into why Era2 and RSL3 treatments yield more opposite responses would be of great interest to the field.”

We agree this is an important point that should be discussed in more detail. In the field of ferroptosis, context-dependent (i.e., cell type-specific) effects are common and multiple groups including our own (Dixon) have published extensively on genes and mechanisms that can lead to differences between erastin2 and RSL3 sensitivity. For example, there are studies showing that the mTOR pathway or the p53 pathway can either prevent or promote ferroptosis, depending on the cell type and/or other currently unknown variables. To address more specifically the differences between Era2 and RSL3 in the context our observed cell-sizedependent response, we have now added more data and discussion. In the Results section we added panel 4B and the following text:

Line 359: “While the upregulation of GSH biosynthesis may promote the resistance of larger cells to ferroptosis, such an upregulation alone cannot explain why larger cells become more resistant to ferroptosis induced by the cystine import inhibitor Era2, but not, for example, by the GPX4 inhibitor RSL3 (Chan et al, 2025) (Figs. 2B, S1B). We found previously that upon mTORC1 inhibition cells can evade cystine deprivation-induced ferroptosis by uptake and catabolism of cysteine-rich extracellular proteins, mostly albumin (Armenta et al, 2022) (Fig. S3C). This process involves albumin degradation in lysosomes, predominantly by cathepsin B (CatB), and subsequent export of cystine from lysosomes to fuel the synthesis of glutathione. Large cells undergo proteome rearrangements similar to those occurring upon mTORC1 inhibition (Zatulovskiy et al, 2022). This suggests that large cells may upregulate CatB expression to bypass the Era2-induced cystine import inhibition via system xc-. To test this hypothesis, we used flow cytometry to measure how the expression of cathepsin B and the system xc- cystine/glutamate transporter SLC7A11 (xCT) scales with cell size (Fig. 4B). We found that SLC7A11 concentration modestly decreases, while CatB concentration significantly increases with cell size (Fig. 4B). This shift in the ratio between SLC7A11 and CatB supports the hypothesis that larger cells may rely less on cystine import via system xc- and thus become more resistant to system xc- inhibition by Era2.”

Additionally, in the Discussion we added the following:

Line 578: “We show that large cells may become resistant specifically to Era2 but not RSL3 through the upregulation of lysosomal function, particularly cathepsin B expression, which enables the uptake and catabolism of cysteine-rich extracellular proteins. A size-dependent shift in the ratio between SLC7A11 and cathepsin B makes large cells less dependent on cystine import via system xc-, and thus, more resistant to Era2. In addition to this, it was reported that RSL3 can induce ferroptosis independently of GPX4 and may target other selenoproteins (DeAngelo et al, 2025; Cheff et al, 2023), which could also contribute to the difference in sizedependent responses to RSL3 and Era2.”

Is the BODIPY-C11 labeling specific to plasma membrane, as suggested by the writing of the authors, or do the results shown integrate signals over all cell membranes?

We thank the reviewer for pointing this out. BODIPY-C11 581/591 stains many membranes in the cell, not just the plasma membrane. We have changed the wording in the manuscript to reflect this.

How exactly is gating done for the flow cytometry samples? Especially when analyzing size-scaling, the results are likely to be sensitive to outliers, such as those seen in Fig 4C (a subpopulation of very low CFSE stained cells). Can the authors clarify their methods and/or display supplementary figures with gating examples?

We have now specified our gating strategy in the Methods section (Line 663) and added a corresponding Supplementary Figure S5.

In Fig 4, total protein staining was used as a control, whereas Fig 5B b-actin was used as a control. Why did the authors rely on different controls approaches for essentially the same measurements? Are these controls comparable?

In our flow cytometry experiments, we consistently use live-cell total protein stain (CFSE) for live cells, and anti-Tubulin immunofluorescent staining for fixed cells, both of which scale in proportion to cell volume and act as a read-out for total cellular protein content (Lanz and Zatulovskiy et al., Mol Cell 2022; Berenson et al. MBoC 2019), which we use to calculate concentrations of other cellular components (analogous to loading controls). In Fig. 5B, betaActin is used as a reference - a protein whose concentration does not change with cell size, as opposed to ACSL4 whose concentration decreases with cell size. In this plot, both ACSL4 and beta-Actin amounts were normalized to alpha-Tubulin, which is analogous to a concentration calculation using loading control. This is now explained in more detail in the Figure legend.

Reviewer #1 (Significance):

I work in the cell size research field, and I am familiar with other related works in this field. My evaluation reflects a specialist's view of this study. Overall, this study will be of a large interest to a small group of specialists, and specific aspects of the work will also gain some interest from broader basic research audiences studying mechanisms of drug responses and ferroptosis in general. However, I do not see this work gaining very broad interest across larger audiences, simply because the field of cell size research is not of broad interest, and this is not a landmark study for the field.

The field of cell size research has long searched for size-dependent functions, as these could help explain why cell size matters. This study is a nice addition to our field, helping establish ferroptosis as a size-dependent function. However, the significance of this work relies on how clearly the authors can establish that their results are cell size rather than cell cycle effects (see major comments above). Should the authors address these concerns, then this study will provide some conceptual and mechanistic insight.

Regarding mechanistic insights, this work is in stark contrast to a recent study about sizedependency of ferroptosis (https://doi.org/10.1016/j.isci.2025.112363), where increased cell size heightened sensitivity to the GPX4 inhibitor RSL3, thus suggesting an opposite conclusion than what the authors observed with the drug Era2. The authors examined this contradiction, and while their results with the drug RSL3 agreed with the recent study, they did not explain why different drug mechanisms yield opposite results. Providing more insights into this discrepancy would increase the impact of this work.

Regardless of the impact of this work, I want to emphasize that I am fully supportive of seeing this work published once the technical concerns have been addressed. Our field will benefit from this work, and this work could catalyze important future research. The general topic studied here has the potential to become very important.

We thank the reviewer for their thoughtful assessment and for supporting publication pending resolution of the technical concerns. We respectfully disagree that our audience is likely narrow: Reviewer #2 noted broad relevance to specialists in cell death/ferroptosis, redox biology, cancer biology, aging, and translational efforts in ferroptosis-based therapies, and Reviewer #3 similarly emphasized both cell size and ferroptosis/cell death communities. We therefore believe the work will be of interest across multiple active fields, particularly because it highlights how cell size heterogeneity can shape drug response.

We agree that the significance hinges on clearly distinguishing cell size from cell-cycle effects, and we have strengthened the corresponding controls/analyses and adjusted language accordingly (see responses to major comments above). We also addressed the reported discrepancy between Era2 and RSL3 size-dependencies by adding new data (Fig. 4B) and expanded discussion. We very much hope that the reviewer appreciates the efforts we have made to strengthen this manuscript and resolve the technical concerns. For these reasons, we believe this work will have an impact on several fields and gain a broad readership.

Reviewer #2:

Zatulovskiy et al. demonstrate that cell size modulates susceptibility to ferroptosis, a form of iron-dependent cell death driven by lipid peroxidation. Using human cell lines (HMEC, HT-1080, RPE-1), the authors examined cell size through FACS sorting, CDK4/6 inhibition and inducible cyclin D1 knockdown. They found that larger cells are more resistant to ferroptosis induced by system xc^-⁻ inhibition (erastin2), but more sensitive to GPX4 inhibition (RSL3), highlighting pathway-specific size dependencies.

Mechanistically, larger cells exhibited:

- Higher glutathione levels, supporting lipid peroxide detoxification

- Increased ferritin expression, promoting iron sequestration

- Lower ACSL4 levels, reducing incorporation of peroxidation-prone lipids

These findings were supported by high-throughput microscopy, flow cytometry (BODIPY-C11 lipid peroxidation assays), and proteomic analyses. The study concludes that cell size influences proteome composition and metabolic capacity, thereby shaping cell death decisions, an insight with implications for aging, cancer, and ferroptosis-based therapies.

Major Comments

(1) Direct evaluation of SLC7A11 abundance and function is needed

The opposite size-dependent effects of erastin2 and RSL3 strongly suggest a role for SLC7A11/system xc^- activity in size-dependent ferroptosis resistance. However, SLC7A11 levels were not quantified due to insufficient peptide detection in the proteomic data. o Direct measurement of SLC7A11 protein levels (immunoblotting or flow cytometry) in small vs large cells would test whether its expression scales with size.

a) Functional perturbation (siRNA/CRISPR knockdown) followed by erastin2 treatment would provide mechanistic validation. o Use of additional SLC7A11 inhibitors (e.g., sulfasalazine, sorafenib) could further test whether the size resistance phenotype is xc^--specific.

We agree that the difference in size-dependent responses to RSL3 and Era2 is an important point that needs further investigation and discussion, as other reviewers also pointed out. To address more specifically the differences between Era2 and RSL3 in the context of cell-sizedependent response, we have now added more data and discussion. In the Results section we added panel 4B measuring SLC7A11 and Cathepsin B scaling with cell size and the following text:

Line 359: “While the upregulation of GSH biosynthesis may promote the resistance of larger cells to ferroptosis, such an upregulation alone cannot explain why larger cells become more resistant to ferroptosis induced by the cystine import inhibitor Era2, but not, for example, by the GPX4 inhibitor RSL3 (Chan et al, 2025) (Figs. 2B, S1B). We found previously that upon mTORC1 inhibition cells can evade cystine deprivation-induced ferroptosis by uptake and catabolism of cysteine-rich extracellular proteins, mostly albumin (Armenta et al, 2022) (Fig. S3C). This process involves albumin degradation in lysosomes, predominantly by cathepsin B (CatB), and subsequent export of cystine from lysosomes to fuel the synthesis of glutathione. Large cells undergo proteome rearrangements similar to those occurring upon mTORC1 inhibition (Zatulovskiy et al, 2022). This suggests that large cells may upregulate CatB expression to bypass the Era2-induced cystine import inhibition via system xc-. To test this hypothesis, we used flow cytometry to measure how the expression of cathepsin B and the system xc- cystine/glutamate transporter SLC7A11 (xCT) scales with cell size (Fig. 4B). We found that SLC7A11 concentration modestly decreases, while CatB concentration significantly increases with cell size (Fig. 4B). This shift in the ratio between SLC7A11 and CatB supports the hypothesis that larger cells may rely less on cystine import via system xc- and thus become more resistant to system xc- inhibition by Era2.”

Additionally, in the Discussion we added the following:

Line 578: “We show that large cells may become resistant specifically to Era2 but not RSL3 through the upregulation of lysosomal function, particularly cathepsin B expression, which enables the uptake and catabolism of cysteine-rich extracellular proteins. A size-dependent shift in the ratio between SLC7A11 and cathepsin B makes large cells less dependent on cystine import via system xc^-, and thus, more resistant to Era2. In addition to this, it was reported that RSL3 can induce ferroptosis independently of GPX4 and may target other selenoproteins (DeAngelo et al, 2025; Cheff et al, 2023), which could also contribute to the difference in sizedependent responses to RSL3 and Era2.”

(2) Functional tests of ferritin contribution to resistance are needed Although elevated ferritin (FTH1/FTL) levels in larger cells represent a strong correlational signal, definitive experimental evidence establishing causality is currently lacking. o Measuring the labile iron pool directly in size-stratified populations would strengthen the link. o Knockdown of FTH1 or FTL could reveal whether ferritin upregulation is necessary for the resistance of large cells to ferroptosis.

We thank the reviewer for raising this point. We have now completed additional experiments, as suggested by the reviewer, and found that iron chelation is unlikely to mediate the sizedependent response to Era2. We have modified the manuscript accordingly and added the following data and discussion to address this point:

Line 296: “The observed increase in ferritin concentration with cell size could therefore lead to additional Fe2+ ion chelation, which in turn would protect large cells from iron-dependent lipid peroxidation and ferroptosis. However, when we measured the concentration of labile intracellular Fe2+ using a fluorescent probe FerroOrange (Hirayama et al, 2020), we did not observe any size-dependent decrease in labile iron concentration (Fig. S2A). Previous work suggests a link between increased sequestration of ferrous iron in lysosomes and resistance to ferroptosis. It was reported that senescent cells, which are also large (Fig. S3A,B), gain resistance to ferroptosis through lysosomal alkalinization and sequestration of ferrous iron in lysosomes (Loo et al, 2025). We therefore tested whether the superscaling of lysosomes observed in large cells (Lanz et al, 2022; You et al, 2025) promotes Era2 resistance through lysosomal iron sequestration. To do this, we stained the cells with the lysosomal iron detection probe Lyso-FerroRed (Saimoto et al, 2025) and measured its scaling using flow cytometry (Fig. S2B). We observed that the amount of Lyso-FerroRed, and therefore, the amount of lysosomal iron, scaled in direct proportion to cell size, just like the total cellular protein content (Fig. S2B). These results indicate that iron chelation by ferritin and its sequestration in lysosomes are unlikely to play a crucial role in size-dependent decrease in Era2 sensitivity.”

(3) Relevance to senescence should be addressed experimentally or explicitly discussed

Given that senescent cells are enlarged and accumulate in aged and tumour tissues, testing senescent models for erastin2 resistance would greatly strengthen the physiological significance.”

We agree that an increase in cell size contributing to the resistance of senescent cells to ferroptosis is intriguing. We have now added a Supplementary Figure S3 and discussion of this point in the manuscript as follows:

Discussion line 552: “…our data suggest that previously reported resistance of senescent cells to ferroptosis can at least partially be due to the increased cell size, a well-established hallmark of senescence.”

Minor Comments

(1) Mechanistic nuance regarding RSL3 should be included

RSL3 has been reported to induce ferroptosis independently of GPX4 (PMID: 37087975, PMID: 40392234) and may target other selenoproteins such as TXNRD1. This nuance would help explain the observed divergence between RSL3 and erastin2 sensitivity across sizes.

We have now added this in the Discussion as suggested by the reviewer (line 583):

“In addition to this, it was reported that RSL3 can induce ferroptosis independently of GPX4 and may target other selenoproteins (DeAngelo et al, 2025; Cheff et al, 2023), which could also contribute to the difference in size-dependent responses to RSL3 and Era2.”

(2) Dynamic range of BODIPY-C11 assays needs commentary

Despite high erastin2 doses, the oxidized BODIPY signal remains close to DMSO levels. The authors should comment on whether this reflects high GSH buffering capacity, probe limitations, or other factors.”

We believe there are both technical (narrow dynamic range of the probe) and biological reasons for the relatively small (2-3 fold) difference in Oxidized-to-Non-oxidized BODIPY-C11 ratios between DMSO and Era2-treated cells. The biological reason is that the cells continue producing GSH until they fully deplete the cystine pool, which happens ~20-24 h after Era2 addition. Once the cystine pool is depleted, the cells very rapidly deplete GSH and initiate cell death. Therefore, there is only a short time window where cells are strongly depleted of GSH before dying. We see this small fraction of cells with a high Oxidized BODIPY-C11 signal in our flow cytometry experiments and in previous microscopy analysis of BODIPY-C11 (Murray et al., Protocol for detection of ferroptosis in cultured cells. STAR Protoc. 2023), but at our chosen time point (20h Era2) most cells are not as bright because we aimed to analyze the population before the onset of widespread cell death.

(3) Western blot for shCycD1 depletion should be included

CycD1 depletion usually causes cells to stop proliferating, which is not the case here. Therefore, depletion must be partial. The level of depletion should be shown by immunblotting.”

The CCND1 manipulation model is extensively characterized in our recent work cited in this manuscript (You et al. (2025), Cell size-dependent mRNA transcription drives proteome remodeling. 2025.10.30.685141 doi:10.1101/2025.10.30.685141). CCND1 shRNA cells do not fully arrest in G0/G1 because the concentration of Cyclin D1 protein in this system is only partially decreased, as the reviewer noted. As a result, the cells have a slightly elongated G1 phase due to a ~30% reduction in Cyclin D1 concentration, but continue to proliferate. The G1 fraction changes from ~70% in wild-type to ~80% in CCND1 shRNA cells.

Reviewer #2 (Significance):

General Assessment: This study presents a mechanistic link between cell size and ferroptosis susceptibility. Using high-throughput microscopy, proteomics, and genetic perturbations across multiple human cell lines, the authors demonstrate that larger cells are more resistant to ferroptosis induced by system xc^- inhibition (erastin2). This resistance is attributed to elevated glutathione production, increased ferritinmediated iron sequestration, and reduced ACSL4-dependent lipid peroxidation. The experimental design is rigorous and multifaceted, with consistent results across cell types and size manipulation methods. While the study is limited to in vitro systems, its conceptual and mechanistic insights lay the groundwork for future in vivo and translational investigations.

Advance: This work is the first to systematically show that cell size directly influences ferroptosis susceptibility via proteome scaling. It reconciles previous findings that large cells are sensitized to GPX4 inhibition (RSL3) by demonstrating that the ferroptosis pathway targeted system xc^- vs GPX4 determines the direction of size-dependent vulnerability. The study provides a conceptual advance by positioning cell size as a regulatory axis in cell death decisions, and a mechanistic advance by identifying size-dependent changes in glutathione metabolism, ferritin levels, and ACSL4 expression.

Audience: This research will be of interest to specialists in cell death, ferroptosis, redox biology, and cancer biology. It also holds relevance for aging researchers and translational scientists exploring ferroptosis-based therapies. The findings may influence how cell size heterogeneity is considered in therapeutic design, particularly in oncology and senescence-targeting strategies.

Field of Expertise: Translational cancer biology, cell cycle regulation, proteomics, therapy resistance, molecular mechanisms of cell death.

We thank Reviewer #2 for their careful and constructive assessment of our manuscript. We were happy that they appreciated the rigor of our multifaceted approach. We are also grateful for their thoughtful perspective on the conceptual and mechanistic advances, and for highlighting the broader relevance of this work to ferroptosis biology, redox regulation, cancer and aging research.

Reviewer #3 (Evidence, reproducibility and clarity):

In this manuscript, Zatulovskiy and colleagues elaborate on their previous work describing cell size-dependent changes in the proteome by investigating whether these changes can be correlated in differences in cell physiology. Using a cleverly-designed high throughput screen, they searched for compounds that differently-sized cells display differential sensitivity towards. Their primary hit, Era2, is involved in the ferroptosis pathway and serves as the starting point for a detailed study of how excess cell size protects cells from ferroptosis-induced cell death via: 1) lower concentrations of ACSL4 (which produces peroxidation-prone PUFAs), 2) increased ferritin concentrations, and 3) increased GSH concentrations.

Overall, the experiments in this manuscript are well-designed and interpreted. It is an extremely well-written manuscript with a clear trajectory of logic. I have only a few major concerns that should be addressed before publication:

We thank Reviewer #3 for their careful reading of the manuscript and for the clear summary of our study and its central findings. We appreciate their positive assessment of the experimental design, interpretation, and overall clarity of the writing and logical flow. We are also grateful for their constructive feedback and take their major concerns seriously; we have addressed each point in detail below.

Major concerns:

(1) In Figure 3E, the authors gate their flow cytometry data using SYTOX so that they are only analyzing live cells. Based on their gating scheme, it seems like there are really a lot of dead cells. Presumably the cells that died were the most sensitive to Era2, so it seems an oversight to discard these cells. Of course, it is not appropriate to analyze dead cells, but this could potentially be solved by using a shorter treatment duration than 24 hours wherein fewer cells die.”

This is a good point. To address it, we have now replaced this panel with a time point where most cells are still alive (20 h, 0.2 µM Era2), as suggested by the reviewer (Fig. 3E,F). This did not change the conclusion that BODIPY-C11 oxidation decreases with cell size.

(2) In Figure 5, are the small, medium, and large bins for ACSL4 KO cells the same as for WT cells? If the ACSL4 KO cells are just bigger to begin with, this could explain why the "small" bin has greater cell survival than the WT small bin. Moreover, is the overlap between the three bins the same in the WT and KO cells?

This is an important point that we now address with data shown in Fig. S4B. We have now added a Supplementary Figure S4B to show the relative size of small, medium, and large WT and ACSL4 KO HMEC cells. As seen from this graph, the ACSL4 KO cells are not bigger than WT cells. Importantly, the fold-range between the small and large FACS-sorted cells is similar (~1.9 to 2-fold).

(3) Loo, et al. Nat Comms 2025 similarly found that senescent cells (which are enlarged) are resistant to ferroptosis using the same inhibitor as the authors. In contrast to the authors, they show that this is due to lysosomal alkalinization and sequestration of ferrous iron in lysosomes. Given that Lanz et al. 2022 found that lysosomal components super-scale with cell size, it seems like this would be an important hypothesis to address. Free lysosomal iron can be easily measured with the LysoRhoNox stain. Loo et al. was able to restore ferroptosis sensitivity in senescent cells using the V-ATPase activator EN6, so it would be important for the authors to address whether this (or similar) treatment would have the same effect in enlarged cells.

This is an excellent point. We have now performed this experiment and added it to the manuscript, as suggested by the reviewer. Based on the Lyso-FerroRed staining (another brand name for the LysoRhoNox probe), we do not see an increase in lysosomal iron sequestration in large cells (Fig. S2B):

Line 301: “Previous work suggests a link between increased sequestration of ferrous iron in lysosomes and resistance to ferroptosis. It was reported that senescent cells, which are also large (Fig. S3A,B), gain resistance to ferroptosis through lysosomal alkalinization and sequestration of ferrous iron in lysosomes (Loo et al, 2025). We therefore tested whether the superscaling of lysosomes observed in large cells (Lanz et al, 2022; You et al, 2025) promotes Era2 resistance through lysosomal iron sequestration. To do this, we stained the cells with the lysosomal iron detection probe Lyso-FerroRed (Saimoto et al, 2025) and measured its scaling using flow cytometry (Fig. S2B). We observed that the amount of Lyso-FerroRed, and therefore, the amount of lysosomal iron, scaled in direct proportion to cell size, just like the total cellular protein content (Fig. S2B). These results indicate that iron chelation by ferritin and its sequestration in lysosomes are unlikely to play a crucial role in size-dependent decrease in Era2 sensitivity.”

Minor concerns:

(1) It would be helpful if this manuscript were re-submitted with line numbers to more easily reference the text.

We have added line numbers for convenience.

(2) In Figure 5A and other figures that reproduce data from Lanz et al. 2022, it would be helpful to have a summary curve for the overall abundance of each protein rather than only the individual peptide curves. These plots (particularly Figure 5A) are difficult to interpret since some peptides were presumably more abundant / measured with higher confidence than others.

We have added the average ACSL4 protein slope line to Fig. 5A.

(3) In Figure 5, the authors show the validation of the ACSL4 KO HT-1080 cell line but not HMEC, even though both are used in this figure. It would be useful to show both. Additionally, the authors switch back and forth between the two cell lines for this figure, and it is not clear why.

We have added the HMEC ACSL4 KO validation Western blot in Fig. S4A.

For the BODIPY oxidation experiment (Fig. 5D), we used HT-1080 instead of HMEC because HT1080 cells are sensitive to lower concentrations of Era2, and therefore, we could better optimize the Era2 concentrations and treatment durations to measure BODIPY oxidation at the time point when most cells are still alive but demonstrate a pronounced oxidized BODIPY signal.

(4) In Figure 5B, the authors use antibody-based staining of ACSL4 and flow cytometry to correlate a loss of ACSL4 expression with increased cell size, validating the proteomics data in Figure 5A. This does not seem like a good way to do this. Firstly, fixing cells with formaldehyde alters their size (is this proportional across differently sized cells? It's impossible to know), which makes it inappropriate to use SSC as a proxy for size in this particular situation. Secondly, the normalization scheme here doesn't make sense. If actin was used as a reference protein, why was tubulin used to normalize ACSL4 abundance? Overall, this seems like a very round-about experiment that could have just been addressed by doing a simple western blot with the four size bins sorted from live cells (as it was in the proteomics). If the issue is that ACSL4 is not detectable by western in the HMEC cells, another solution would be plating the live, sorted bins on coverslips and measuring by IF (or using the HT-1080 cells).

We prefer IF flow cytometry to Western blotting for protein scaling analysis because it is more quantitative and provides cell size and protein content information for each individual cell. While in principle, different-sized cells might change their size differently during fixation, the cells that were larger or smaller prior to the fixation remain larger or smaller after fixation as well.

Therefore, the SSC measurement after fixation still provides reliable information on size ranking, even if SSC does not perfectly linearly scale with cell volume. We do not use the SSC information to calculate protein concentrations here. Instead, we divide the amount of our protein of interest in the cell by the amount of constitutively-expressed Tubulin, which acts as an analogue of a loading control in this experiment. In Fig. 5B, both ACSL4 and Actin were normalized to Tubulin to estimate their concentrations. Actin is used just as a reference protein to show how the concentration of a perfectly scaling protein remains constant across cell size, as opposed to the sub-scaling ACSL4. Tubulin in this case was used as a proxy for total cellular protein content, which scales linearly in proportion to cell volume. This approach for determining the scaling behaviors of different proteins was previously validated in Lanz et al., Mol Cell 2022.

(5) In Figure 5E/5F, the authors pre-arrest the cells in G1 with palbociclib before size-sorting them. The pre-arrest is not done in other experiments using this cell line for sizesorting, so it would be important for the authors to comment on why this was done for this experiment but not others.”

As we found in Fig. 2B-E, the cell cycle has confounding effects on size-dependent ferroptosis susceptibility measurements (as discussed in detail in our response to the first major point of Reviewer #1 above). Briefly, to avoid these confounding effects and isolate the effects of cell size from the effects of the cell cycle, we pre-synchronized the cells with 24 h treatment with palbociclib in Fig. 5E,F. This is now better clarified in the text, as follows:

Line 456: “In this experiment, we synchronized cells in G1 phase using palbociclib prior to cell sorting and also incubated the sorted cells in the presence of palbociclib during Era2 treatment to isolate cell size effects from the previously observed confounding effects of the cell cycle on ferroptosis (Fig. 2B,E).”

(6) Conceptually, it is difficult for me to understand why large cell size sensitizes cells to GPX4 inhibition but confers resistance to Era2 treatment. Particularly given the pathway described in Figure 3A, I am having trouble understanding why these would convey such opposing phenotypes. Shouldn't the extra ferritin in the bigger cells also help them cope with GPX4 inhibition if, as the authors state in the discussion, the increased sensitivity to the GPX4 inhibitor is reported to be mediated by (among other things) iron accumulation? A deeper discussion of this seeming-incongruity would be helpful for contextualizing the broader role of cell size in determining ferroptosis sensitivity.

We agree this is an important point, which was also raised by the other reviewers. As such, we note that context-dependent (i.e., cell type-specific) effects are common in the ferroptosis field, and multiple groups including our own (Dixon) have published extensively on genes and mechanisms that can lead to differences between erastin2 and RSL3. For example, there are studies showing that the mTOR pathway or the p53 pathway can both prevent and promote ferroptosis, depending on the cell type or some other hidden variable.

To better address the differences between Era2 and RSL3 in the context of the cell-sizedependent response, we have now added more data and discussion. In the Results section we added panel 4B and the following text:

Line 359: “While the upregulation of GSH biosynthesis may promote the resistance of larger cells to ferroptosis, such an upregulation alone cannot explain why larger cells become more resistant to ferroptosis induced by the cystine import inhibitor Era2, but not, for example, by the GPX4 inhibitor RSL3 (Chan et al, 2025) (Figs. 2B, S1B). We found previously that upon mTORC1 inhibition cells can evade cystine deprivation-induced ferroptosis by uptake and catabolism of cysteine-rich extracellular proteins, mostly albumin (Armenta et al, 2022) (Fig. S3C). This process involves albumin degradation in lysosomes, predominantly by cathepsin B (CatB), and subsequent export of cystine from lysosomes to fuel the synthesis of glutathione. Large cells undergo proteome rearrangements similar to those occurring upon mTORC1 inhibition (Zatulovskiy et al, 2022). This suggests that large cells may upregulate CatB expression to bypass the Era2-induced cystine import inhibition via system xc-. To test this hypothesis, we used flow cytometry to measure how the expression of cathepsin B and the system xc- cystine/glutamate transporter SLC7A11 (xCT) scales with cell size (Fig. 4B). We found that SLC7A11 concentration modestly decreases, while CatB concentration significantly increases with cell size (Fig. 4B). This shift in the ratio between SLC7A11 and CatB supports the hypothesis that larger cells may rely less on cystine import via system xc- and thus become more resistant to system xc- inhibition by Era2.”

Additionally, in the Discussion we added the following:

Line 578: “We show that large cells may become resistant specifically to Era2 but not RSL3 through the upregulation of lysosomal function, particularly cathepsin B expression, which enables the uptake and catabolism of cysteine-rich extracellular proteins. A size-dependent shift in the ratio between SLC7A11 and cathepsin B makes large cells less dependent on cystine import via system xc-, and thus, more resistant to Era2. In addition to this, it was reported that RSL3 can induce ferroptosis independently of GPX4 and may target other selenoproteins (DeAngelo et al, 2025; Cheff et al, 2023), which could also contribute to the difference in sizedependent responses to RSL3 and Era2.”

In some ways this title seems to run at cross purposes to your argument that we should discuss Mars and Squire as disciplined, experimental artists and not just remember them as colorful lesbians. Shouldn't your title then put the spotlight on the art and not on queer identities? Maybe "Alternative Queer Artistries"?

There is no entry here for Ethel Mars, Artsy, so I am unable to locate in the Works Cited the various paintings by Mars you included in the story.

Their paintings are gorgeous, but why do two different paintings have the same caption: Ethel Mars, Artsy? Having never seen these sketches before, I crave more immediately accessible information about titles, dates, locations, etc...

I have a hard time making sense of this quotation without better framing/ sandwiching. Who is Sherwood? and what American Girl is Sherwood talking about?

Here again, this road sign doesn't adequately signal the blend of biography, close reading, and artistry that will unfold beyond the button. Embed those videos so we can't miss them!

ohmigosh, this is the most ingenious, engaging reimagining of humanities scholarship I have ever encountered! I love it so much! I wish you would embed the youtube video into your storymap rather than hiding it behind a button that reader's might skip (the button also doesn't explain what we're about to see -- I thought it was simply a link to the painting).

Changes in font size can be distracting unless used consistently to create clear visual hierachies and meaningful distinctions.

Brilliant intervention!

delete this comma

See George Chauncey's social history, "Gay New York" for discussion of the history of the term. He suggests that the queer meaning was circulating in the early 20th century,, at least in queer subcultures. The Oxford English Dictionary (available online via our library) includes examples of the first recorded usage of any meaning. They actually cite Stein's portrait as the first text to use "gay" to mean homosexual! Stein changed the language! https://www.oed.com/dictionary/gay_adj?tab=meaning_and_use#3287607.

Bold and dramatic insight! I'm intrigued! What you mean by linguistic impact? And who or what did these portraits have an impact on?

Your exposition is bold and engaging. Here, I think it would be more accurate to say: Stein created portraits of modernist giants such as...

I don't think these artists "possess" their abstract word portraits, since, as George Bornstein reminds us, literary texts don't exist in any one place or materialization.

I wonder if Hutcheon's discursive definition of art always existing in a sea of past references renders the notion of "a truth" in flux. Maybe it would make more sense to say that comparing two or more works puts them in conversation, and that the scholar who compares them joins the conversation. But what happens when scholars contribute their own artistry to the conversation?

I'm just not convinced that there is "a truth" or "the truth" embedded in any text, and I don't think that's a loss, but a gain for the way we exist in conversation with texts from the past.

The breakdown of various finding tools was especially helpful, as the purpose of each tool is explained clearly and practically. Organizing the chapter by tool type makes it easier to identify which resources are best for different stages of research.

It was cool to check out the timeline of these changes, and I had no idea about this evolution before.

The timeline in this part was super helpful because it shows how research tools have changed over the years. Seeing the history helps make sense of today's search systems and puts our current research methods into perspective. It was really interesting to watch how these tools have evolved.

This clarification really matters because people sometimes think that just because they find info in a database or search engine, it’s automatically trustworthy. The section does a great job of showing that finding information and judging its reliability are two different things.

This approach is useful for reviewing the reasoning behind material choices. It can help prevent accidental plagiarism. The Cornell Notes strategy encourages writers to first process information in their own words. This practice can make paraphrasing feel more natural and original.

The use of signal phrases is often taught merely as citation tools, ignoring their role in creating a cohesive writing structure. These examples demonstrate how signal phrases can enhance the flow and readability of academic writing.

This helps in finding plagiarism; making it much easier to recognize in practice.

Having these video was very helpful for refreshing how the notes are done.

x

object of discovery

post critical

does not sound right to me

is reasonable

in the sense that it satisfies all the conditions for its existence

There is a third

future sense

gestalti wholistic

future oriented!

Left hemisphsre primacy of the right

it is not a formal demand

The golden rule is generative

in harmony with the world

not knowing ut awareness

knowing

beyond justified true belief

anamnesis, knowing that a path exists to what is not yet acomplished

pro not anamnesis

There is lot more to it than what is the case

the real question does what we follow maximises future choices or not

Future perfect understanding

transcendtal analytic

vs

it works

higher awareness awareness becomming aware of itself

like drawing hands

I get the reasons why you rail against the absolute

the in betweenness does speak to human potential to ascend to local absolutes which are self-hosting once a new level of self-created complexity is ascended to, growth of undestanding if you like and civilization levels these are worlds to themsleves that we create to inhabit

not becoming divine but closer to it

language itself is a self correcting self extending strange loops of loops

Consider the trefoil no beginning no end yet finite self-enclosing

??

there is finite loops autopoiesys. Life itself is an example It is not infinite, but I would call a local absolute a local maxima

re hill climbing

human potential is such

a humanistic account that does not talk about 'human potential' to me is puzzling

what a God foresaken conception

love of wisdom

eclipsed the

Wisdom in Love

understand what is the case

antroposopia?

perniscious indeed

yes

absolute understanding

destroys what's human

local absolutes are real human spirit is one of those

most important question

thinking

very strong statement

to me this should be the final paragraph of the Introduction. Go ahead and enumerate hypothesis and consider the order you state them and follow that same pattern throughout the paper.

survival?

I assume we see this in Fig 3?

increase font in Fig 3

now jumping back to first person..

Survival?

same - do no lead with ..

also Embryo count sounds awkward --

do not state with name of plot .. start with Embryo count at .... then describe what box plot represents

Do not lead with method, start with Polyvinyl chloride

be explicit - in RNA there was pooling- do these correspond to those - how many at each stage- probably best to lead with in 2.9.1 DNA Ampl

is it "~"?

The switch here to first person is a bit abrupt

I would doi the repo with Zenodo then cite as reference.

a few spacing issues

remind reader what these ar

I think this might be Fig 11- but also just explicitly write in text how pooling was done

uno scarico di responsabilità

Vale solo per le foto, se chiamati media sembra che sia disponibile anche per i video. C'è un errore di UI nello screen e un campo non è in fill

Refuso

nel

Invertire ordine delle immagini in basso per seguire lo stesso ordine di come le descrivi

Farei un elenco numerato: 1. Posizionare il punto cliccando... 2. Definire l'orientamento del Waypoint muovendo...

...sua stazione di ricarica

...una missione non ancora eseguita

No nomi di componenti che conosciamo solo noi designer

System 1 uses heuristics

Janie first marriage was with Logan and the wedding took place in nanny’s parlor of Saturday evening

60 layers of land

Making Memorials to the Future Open Access Christine Sylvester Global Studies Quarterly, Volume 3, Issue 4, October 2023, ksad065, https://doi.org/10.1093/isagsq/ksad065

https://pbswisconsin.org/watch/pov/maya-lin-a-strong-clear-vision-7v5tab/

https://news.yale.edu/2012/11/26/designer-911-memorial-speak-saybrook

The continuous downward flow of water symbolizes grief without resolution. Water disappears into a void that cannot be filled, rhetorically representing permanent loss. The repetition also creates meditative rhythm, encouraging contemplation rather than political interpretation.

https://www.researchgate.net/publication/354640770_Enshrining_Terror_for_the_Nation_Affect_and_Nationalism_at_the_National_September_11_Memorial_and_Museum

"The 9/11 and Vietnam Veterans Memorials: Interpreting Experience in the Landscape of Loss" https://ecommons.cornell.edu/entities/publication/0aa6f356-62e8-4eb8-ad4c-273dfebb5dce/full

"Making of the Vietnam Veterans Memorial" https://daily.jstor.org/the-making-of-the-vietnam-veterans-memorial/

Dans ce témoignage, je trouve intéressant le contraste entre la profondeur des échanges décrits et la difficulté à retrouver cette connexion dans la réalité. Ca me fait me demander si l’on peut parfois tomber amoureux autant (voire plus) d’une relation imaginée et de ce qu’elle représente que de la personne elle-même.

Cette idée de mieux comprendre ses attentes personnelles me parle pas mal. J’ai trouvé intéressant le fait que la réflexion dépasse ici la simple question des sites de rencontres pour revenir à quelque chose de plus profond : la connaissance de soi. En revanche, je me demande si nous savons toujours précisément ce que nous recherchons avant une rencontre (en ligne ou hors ligne). J’ai plutôt l’impression que certaines choses se découvrent progressivement, au fil des expériences et au contact de l’autre. On pense parfois savoir exactement ce que l’on veut, puis une rencontre vient complètement bousculer nos critères ou nos certitudes. En tous cas j'apprécie l'idée de ne pas forcément avoir besoin d'une checklist pour trouver l'amour.

La personne semble prendre conscience qu’elle ne recherche pas seulement une personne réelle mais aussi une représentation idéalisée construite à partir de ses attentes personnelles. Cela renforce l’idée que le fantasme amoureux ne naît pas nécessairement du numérique mais peut être amplifié par celui-ci.

L’idée d’un décalage entre l’image que l’on construit et la réalité me semble pertinente. En revanche, la déception apparaît ici comme une conséquence presque inévitable, ce qui me paraît être une généralisation importante. L’imaginaire et l’idéalisation existent aussi dans des rencontres hors ligne et peuvent parfois être renforcés, mais aussi contredits ou enrichis par la rencontre réelle.

L’utilisation du terme « hystérisent » attire mon attention. Ce choix de vocabulaire semble particulièrement fort et peut orienter la perception du lecteur. Je me demande si cette interprétation repose sur des recherches scientifiques ou davantage sur une observation clinique personnelle.

L’opposition entre comportements masculins et féminins paraît ici très catégorique. Existe-t-il des recherches permettant de soutenir ces différences ou risque-t-on de renforcer certains stéréotypes de genre ?

Je trouve intéressant le lien établi entre abondance de choix et quête de l’idéal. Lorsqu’on a le sentiment qu’il existe toujours quelqu’un de potentiellement plus compatible, cela peut effectivement nourrir l’idée qu’il est possible d’atteindre une forme de perfection relationnelle.En revanche, l’expression « jeter l’autre » est particulièrement forte et véhicule une représentation assez négative des relations contemporaines. Quelles données empiriques derrière le fait de jeter plus aujourd'hui à cause des rencontres en ligne ?

eLife Assessment

This study thoroughly assesses tactile acuity on women's breasts, for which no dependable data currently exists. The study provides two important contributions, by convincingly showing that tactile acuity on the breast is poor in comparison to other body parts, and that acuity is worst in larger breasts, indicating that the number of tactile sensors is fixed. This study will be of interest to the broader community of touch, as well as those interested in breast reconstruction and sexual function.

Reviewer #1 (Public review):

[Editors' note: this version has been assessed by the Senior Editor without further input from the original reviewers. The authors have moderated their claims and discussed the limitations of their experimental design more transparently. The previous reviews are included for reference.]

Comments on previous version:

The authors investigated tactile spatial perception on the breast using discrimination, categorization, and direct localization tasks. They reach four main conclusions:

(1) The breast has poor tactile spatial resolution.<br /> This conclusion is based on comparing just noticeable differences, a marker of tactile spatial resolution, across four body regions, two on the breast. The data compellingly support the conclusion; the study outshines other studies on tactile spatial resolution that tend to use problematic measures of tactile resolution, such as two-point-discrimination thresholds. The result will interest researchers in the field and possibly in other fields due to the intriguing tension between the finding and the sexually arousing function of touching the breast.

The manuscript incorrectly describes the result as poor spatial acuity. Acuity measures the average absolute error, and acuity is good when response biases are absent. Precision relates to the error variance. It is common to see high precision with low acuity or vice versa. Just noticeable differences assess precision or spatial resolution, while points of subjective equality evaluate acuity or bias. Similar confusions between these terms appear throughout the manuscript.

A paragraph within the next section seems to follow up on this insight by examining the across-participant consistency of the differences in tactile spatial resolution between body parts. To this aim, pairwise rank correlations between body sites are conducted. This analysis raises red flags from a statistical point of view. 1) An ANOVA and its follow-up tests assume no variation in the size of the tested effect but varying base values across participants. Thus, if significant differences between conditions are confirmed by the original statistical analysis, most participants will have better spatial resolution in one condition than the other condition, and the difference between body sites will be similar across participants. 2) Correlations are power-hungry, and non-parametric tests are power-hungry. Thus, the number of participants needed for a reliable rank correlation analysis far exceeds that of the study. In sum, a correlation should emerge between body sites associated with significantly different tactile JNDs; however, these correlations might only be significant for body sites with pronounced differences due to the sample size.

(2) Larger breasts are associated with lower tactile spatial resolution<br /> This conclusion is based on a strong correlation between participants' JNDs and the size of their breasts. The depicted correlation convincingly supports the conclusion. The sample size is below that recommended for correlations based on power analyses, but simulations show that spurious correlations of the reported size are extremely unlikely at N=18. Moreover, visual inspection rules out that outliers drive these correlations. Thus, they are convincing. This result is of interest to the field, as it aligns with the hypothesis that nerve fibers are more sparsely distributed across larger body parts.

(3) The nipple is a unit<br /> The data do not support this conclusion. The conclusion that the nipple is perceived as a unit is based on poor tactile localization performance for touches on the nipple compared to the areola. The problem is that the localization task is a quadrant identification task with the center being at the nipple. Quadrants for the areola could be significantly larger due to the relative size of the areola and the nipple; the results section seems to suggest this was accounted for when placing the tactile stimuli within the quadrants, but the methods section suggests otherwise. Additionally, the areola has an advantage because of its distance from the nipple, which leads to larger Euclidean distances between the centers of the quadrants than for the nipple. Thus, participants should do better for the areola than for the nipple even if both sites have the same tactile resolution.

To justify the conclusion that the nipple is a unit, additional data would be required. 1) One could compare psychometric curves with the nipple as the center and psychometric curves with a nearby point on the areola as the center. 2) Performance in the quadrant task could be compared for the nipple and an equally sized portion of the areola and tactile locations that have the same distance to the border between quadrants in skin coordinates. 3) Tactile resolution could be directly measured for both body sites using a tactile orientation task with either a two-dot probe or a haptic grating.

Categorization accuracy in each area was tested against chance using a Monte Carlo test, which is fine, though the calculation of the test statistic, Z, should be reported in the Methods section, as there are several options. Localization accuracies are then compared between areas using a paired t-test. It is a bit confusing that once a distribution-approximating test is used, and once a test that assumes Gaussian distributions when the data is Bernoulli/Binomial distributed. Sampling-based and t-tests are very robust, so these surprising choices should have hardly any effect on the results.

A correlation based on N=4 participants is dangerously underpowered. A quick simulation shows that correlation coefficients of randomly sampled numbers are uniformly distributed at such a low sample size. This likely spurious correlation is not analyzed, but quite prominently featured in a figure and discussed in the text, which is worrisome.

(4) Localization of tactile events on the breast is biased towards the nipple<br /> The conclusion that tactile percepts are drawn toward the nipple is based on localization biases for tactile stimuli on the breast compared to the back. Unfortunately, the way participants reported the tactile locations introduces a major confound. Participants indicated the perceived locations of the tactile stimulus on 3D models of these body parts. The nipple is a highly distinctive and cognitively represented landmark, far more so than the scapula, making it very likely that responses were biased toward the nipple regardless of the actual percepts. One imperfect but better alternative would have been to ask participants to identify locations on a neutral grey patch and help them relate this patch to their skin by repeatedly tracing its outline on the skin.

Participants also saw their localization responses for the previously touched locations. This is unlikely to induce bias towards the nipple, but it renders any estimate of the size and variance of the errors unreliable. Participants will always make sure that the marked locations are sufficiently distant from each other.

The statistical analysis is again a homebrew solution and hard to follow. It remains unclear why standard and straightforward measures of bias, such as regressing reported against actual locations, were not used.

Null-hypothesis significance testing only lets scientists either reject the null hypothesis or not. The latter does NOT mean the Null hypothesis is true, i.e., it can never be concluded that there is no effect. This rule applies to every NHST test. However, it raises particular concerns with distribution tests. The only conclusion possible is that the data are unlikely from a population with the tested distribution; these tests do not provide insight into the actual distribution of the data, regardless of whether the result is significant or not.

Reviewer #2 (Public review):

Summary:

The authors tested tactile acuity on the breast of females using several tasks.

Results:

Tactile acuity, assessed by just-noticeable differences in judging whether a touch was above or below a comparison stimulus, was lower on both the lateral and medial breast than on the hand and back. Acuity also scaled inversely with breast size, echoing earlier findings that larger hands exhibit lower acuity, presumably because a similar number of tactile receptors must be distributed over larger or smaller body surfaces. Observing this principle in the breast as on the hand strengthens the view that fixed innervation is a general organizing principle of the tactile system. Both methodology and analysis appear sound.

Most participants were unable to localize touch to a specific quadrant of the nipple, suggesting it is perceived as a single tactile unit. However, the study does not address whether touches to the nipple and areola are confused; conceptualizing the nipple as a perceptual (landmark) unit would suggest that such confusion should not take place. Aside from this limitation, the methodology and analysis appear sound.

Absolute touch localization, assessed by asking participants to indicate locations on a 3D rendering of their own torso, revealed a bias toward the nipple. The authors interpret this as evidence that the nipple serves as a landmark attracting perceived touch. However, as reviewers noted during review, alternative explanations cannot be fully ruled out: because the stimulus array was centered on the nipple, the observed bias may stem from stimulus distribution rather than landmark status. Aside from this caveat, the methodology and analysis appear sound.

Overall assessment:

The study offers a welcome exception to the prevailing bias in tactile research that limits investigation to the hand and arm. Its support for the fixed innervation hypothesis and its suggestion that the nipple may serve as a potential landmark-though requiring further scrutiny-illustrate the value of extending research to other body regions. By employing multiple tasks, the authors address several key aspects of tactile perception and create links to earlier findings.

Author response:

The following is the authors’ response to the previous reviews

Public Reviews:

Reviewer #1 (Public review):

The manuscript incorrectly describes the result as poor spatial acuity. Acuity measures the average absolute error, and acuity is good when response biases are absent. Precision relates to the error variance. It is common to see high precision with low acuity or vice versa. Just noticeable differences assess precision or spatial resolution, while points of subjective equality evaluate acuity or bias. Similar confusions between these terms appear throughout the manuscript.

While I do not agree with the reviewer's usage of the word “acuity” and a cursory Google search does not agree with the provided definition, I have replaced acuity with precision as appropriate to improve clarity.

A paragraph within the next section seems to follow up on this insight by examining the across-participant consistency of the differences in tactile spatial resolution between body parts. To this aim, pairwise rank correlations between body sites are conducted. This analysis raises red flags from a statistical point of view. 1) An ANOVA and its follow-up tests assume no variation in the size of the tested effect but varying base values across participants. Thus, if significant differences between conditions are confirmed by the original statistical analysis, most participants will have better spatial resolution in one condition than the other condition, and the difference between body sites will be similar across participants. 2) Correlations are power-hungry, and non-parametric tests are power-hungry. Thus, the number of participants needed for a reliable rank correlation analysis far exceeds that of the study. In sum, a correlation should emerge between body sites associated with significantly different tactile JNDs; however, these correlations might only be significant for body sites with pronounced differences due to the sample size.

We have entirely removed this result from both the text and supplement.

The data do not support this conclusion. The conclusion that the nipple is perceived as a unit is based on poor tactile localization performance for touches on the nipple compared to the areola. The problem is that the localization task is a quadrant identification task with the center being at the nipple. Quadrants for the areola could be significantly larger due to the relative size of the areola and the nipple; the results section seems to suggest this was accounted for when placing the tactile stimuli within the quadrants, but the methods section suggests otherwise. Additionally, the areola has an advantage because of its distance from the nipple, which leads to larger Euclidean distances between the centers of the quadrants than for the nipple. Thus, participants should do better for the areola than for the nipple even if both sites have the same tactile resolution.

We agree with this interpretation and have updated the language throughout.

Categorization accuracy in each area was tested against chance using a Monte Carlo test, which is fine, though the calculation of the test statistic, Z, should be reported in the Methods section, as there are several options. Localization accuracies are then compared between areas using a paired t-test. It is a bit confusing that once a distribution-approximating test is used, and once a test that assumes Gaussian distributions when the data is Bernoulli/Binomial distributed. Sampling-based and t-tests are very robust, so these surprising choices should have hardly any effect on the results.

Excellent point. We have replaced the paired t-test with a signed rank test and added text to the methods to expand upon this.

A correlation based on N=4 participants is dangerously underpowered. A quick simulation shows that correlation coefficients of randomly sampled numbers are uniformly distributed at such a low sample size. This likely spurious correlation is not analyzed, but quite prominently featured in a figure and discussed in the text, which is worrisome.

We have removed this panel to reduce this concern.

The conclusion that tactile percepts are drawn toward the nipple is based on localization biases for tactile stimuli on the breast compared to the back. Unfortunately, the way participants reported the tactile locations introduces a major confound. Participants indicated the perceived locations of the tactile stimulus on 3D models of these body parts. The nipple is a highly distinctive and cognitively represented landmark, far more so than the scapula, making it very likely that responses were biased toward the nipple regardless of the actual percepts. One imperfect but better alternative would have been to ask participants to identify locations on a neutral grey patch and help them relate this patch to their skin by repeatedly tracing its outline on the skin.

While I wholeheartedly agree with the sentiments of the reviewer, in our experience performing these tests across many women we have found that the variability of the morphology of the breast makes it incredibly hard for women to perform this task in the way the reviewer is describing. Consequently, there is likely no perfect version of the task. That said, we have endeavored to acknowledge the limitations of the approach in the discussion.

Participants also saw their localization responses for the previously touched locations. This is unlikely to induce bias towards the nipple, but it renders any estimate of the size and variance of the errors unreliable. Participants will always make sure that the marked locations are sufficiently distant from each other.

I again respectfully disagree with this interpretation. If the participants were to always make sure marked locations were sufficiently distant from each other then the degree of error and bias would be similar between regions given that the visual pattern would be almost identical. As this is not true in the data, I disagree with the premise, though we hope the changes to the discussion acknowledge limitations with the data collection method.

Null-hypothesis significance testing only lets scientists either reject the null hypothesis or not. The latter does NOT mean the Null hypothesis is true, i.e., it can never be concluded that there is no effect. This rule applies to every NHST test. However, it raises particular concerns with distribution tests. The only conclusion possible is that the data are unlikely from a population with the tested distribution; these tests do not provide insight into the actual distribution of the data, regardless of whether the result is significant or not.

Thank you for this comment. We have updated the language to make it explicit that we do not mean to imply failing to deviate from the Null distribution does not mean that they are in fact Null in nature.

Recommendations for the authors:

Reviewer #2 (Recommendations for the authors):

I am wondering whether the interpretation of "the nipple as a sensory unit" is also supported by localization performance as reported in the analysis around Fig. 3 and supplementary Fig. 2. I cannot really see the error lines in that figure, and cannot tell whether any of the touches were on the nipple proper. Specifically I am wondering whether touch to the nipple is reliably attributed to the nipple, and touch to the areola to the areola, or whether confusion exists between the two. The description of the nipple as a sensory unit implies reliable attribution of touch to the respective area. Also the discussion (lines 309ff) is ambiguous about this.

Thank you for this comment. We have removed language about the nipple being a unit and reframed the text in the discussion. We have also clarified that touches were indeed on the nipple.

typos etc.

lines 68-71 - implied causality is not backed up by evidence and could be the other way around than stated here

line 82 grammar is inconsistent

lines 199-200, "on the nipple" occurs twice

Thank you for catching these. We have addressed the typos and grammar. We have also added a citation to the sentence where this exact hypothesis is stated. We have also relaxed the language to imply it is indeed a hypothesis.

He lives in the backyard

She finally saw peace in her life

eLife Assessment

This important work demonstrates the role of physically linking the core and CTD kinase modules of TFIIH via separate domains of subunit Tfb3 in confining RNA Polymerase II Serine 5 CTD phosphorylation to promoter regions of transcribed genes in budding yeast. The main findings, resulting from analyses of viable Tfb3 mutants in which the linkage between TFIIH core and kinase modules has been severed, are supported by solid evidence from in vitro and in vivo experiments. The new findings raise the intriguing possibility that the Tfb3-mediated connection between core and kinase modules of TFIIH is an evolutionary addition to an ancestral state of physically unconnected enzymes.

Reviewer #1 (Public review):

[Editors' note: this version has been assessed by the Reviewing Editor without further input from the original reviewers. The authors have addressed the comments raised in the previous rounds of review.]

Giordano et al. demonstrate that yeast cells expressing separated N- and C-terminal regions of Tfb3 are viable and grow well. Using this creative and powerful tool, the authors effectively uncouple CTD Ser5 phosphorylation at promoters and assess its impact on transcription. This strategy is complementary to previous approaches, such as Kin28 depletion or the use of CDK7 inhibitors. The results are largely consistent with earlier studies, reinforcing the importance of the Tfb3 linkage in mediating CTD Ser5 phosphorylation at promoters and subsequent transcription.

Notably, the authors also observe effects attributable to the Tfb3 linker itself, beyond its role as a simple physical connection between the N- and C-terminal domains. These findings provide functional insight into the Tfb3 linker, which had previously been observed in structural studies but lacked clear functional relevance. Overall, I am very positive about the publication of this manuscript.

Reviewer #2 (Public review):

Summary:

This work advances our understanding of how TFIIH coordinates DNA melting and CTD phosphorylation during transcription initiation. The finding that untethered kinase activity becomes "unfocused," phosphorylating the CTD at ser5 throughout the coding sequence rather than being promoter-restricted, suggests that the TFIIH Core-Kinase linkage not only targets the kinase to promoters but also constrains its activity in a spatial and temporal manner.

Strengths:

The experiments presented are straightforward and the model for coupling initiation and CTD phosphorylation and for evolution of these linked processes are interesting and novel. The results have important implications for the regulation of initiation and CTD phosphorylation.

Reviewer #3 (Public review):

Summary:

Eukaryotic gene transcription requires a large assemblage of protein complexes that govern the molecular events required for RNA Polymerase II to produce mRNAs. One of these complexes, TFIIH, comprises two modules, one of which promotes DNA unwinding at promoters, while the other contains a kinase (Kin28 in yeast) that phosphorylates the repeated motif at the C-terminal domain (CTD) of the largest subunit of Pol II. Kin28 phosphorylation of Ser5 in the YSPTSPS motif of the CTD is normally highly localized at promoter regions, and marks the beginning of a cycle of phosphorylation events and accompanying protein association with the CTD during the transition from initiation to elongation.

The two modules of TFIIH are linked by Tfb3. Tfb3 consists of two globular regions, an N-terminal domain that contacts the Core module of TFIIH and a C-terminal domain that contacts the kinase module, connected by a linker. In this paper, Giordano et al. test the role of Tfb3 as a connector between the two modules of TFIIH in yeast. They show that while no or very slow growth occurs if only the C-terminal or N-terminal region of Tfb3 is present, near normal growth is observed when the two unlinked regions are expressed. Consistent with this result, the separate domains are shown to interact with the two distinct TFIIH modules. ChIP experiments show that the Core module of TFIIH maintains its localization at gene promoters when the Tfb3 domains are separated, while localization of the kinase module, and of Ser5 phosphorylation on the CTD of Pol II, is disrupted. Finally, the authors examine the effect of separating the Tfb3 domains on another function of TFIIH, namely nucleotide excision repair, and find little or no effect when only the N-terminal region of Tfb3 or the two unlinked domains are present.

Strengths:

Experiments involving expression of Tfb3 domains in yeast are well-controlled and the data regarding viability, interaction of the separate Tfb3 domains with TFIIH modules, genome-wide localization of the TFIIH modules and of phosphorylated Ser5 CTDs, and of effects on NER, are convincing. The experiments are consistent with current models of TFIIH structure and function and support a model in which Tfb3 tethers the kinase module of TFIIH close to initiation sites to prevent its promiscuous action on elongating Pol II.

Author response:

The following is the authors’ response to the previous reviews

eLife Assessment

This important work demonstrates the role of physically linking the core and CTD kinase modules of TFIIH via separate domains of subunit Tfb3 in confining RNA Polymerase II Serine 5 CTD phosphorylation to promoter regions of transcribed genes in budding yeast. The main findings, resulting from analyses of viable Tfb3 mutants in which the linkage between TFIIH core and kinase modules has been severed, are supported by solid evidence from in vitro and in vivo experiments. The new findings raise the intriguing possibility that the Tfb3-mediated connection between core and kinase modules of TFIIH is an evolutionary addition to an ancestral state of physically unconnected enzymes.

After consultation with the referees, we would like to suggest that you insert text into the RESULTS section acknowledging two limitations of your findings remaining in the revised manuscript, as follows:

(i) It remains possible that Kin28 abundance was reduced by splitting Tfb3, which could be a factor in reducing its occupancies at gene promoters.

In response, the paper now contains the following sentence:

“Kin28 levels in extracts were below the limit of detection for our antibody, so we cannot rule out that the drop in ChIP signal is partly due to reduced Kin28 levels in the split Tfb3 strains. However, the viability of the cells (Figure 2) and the Tfb3-TAP purifications (Figure 3) argue against a complete loss of Kin28.”

(ii) Lower than wild-type expression of the Tfb3 truncations might contribute to their mutant phenotypes shown in Figs. 2 & 5.

In response, the paper now contains the following sentence:

“There was some variation in protein expression levels (Figure 3A, left panel, lanes 1-4), and reduced levels of the split Tfb3 may contribute to the slow growth phenotypes.”

Public Reviews:

Reviewer #1 (Public review):

Giordano et al. demonstrate that yeast cells expressing separated N- and C-terminal regions of Tfb3 are viable and grow well. Using this creative and powerful tool, the authors effectively uncouple CTD Ser5 phosphorylation at promoters and assess its impact on transcription. This strategy is complementary to previous approaches, such as Kin28 depletion or the use of CDK7 inhibitors. The results are largely consistent with earlier studies, reinforcing the importance of the Tfb3 linkage in mediating CTD Ser5 phosphorylation at promoters and subsequent transcription.

Notably, the authors also observe effects attributable to the Tfb3 linker itself, beyond its role as a simple physical connection between the N- and C-terminal domains. These findings provide functional insight into the Tfb3 linker, which had previously been observed in structural studies but lacked clear functional relevance. Overall, I am very positive about the publication of this manuscript and offer a few minor comments below that may help to further strengthen the study.

We appreciate the reviewer’s positive assessment of our work and suggestions for improvement.

Page 4 PIC structures show the linker emerging from the N-terminal domain as a long alpha-helix running along the interface between the two ATPase subunits, followed by a turn and a short stretch of helix just N-terminal to a disordered region that connects to the C-terminal region (see schematic in Fig. 1A).

The linker helix was only observed in the poised PIC (Abril-Garrido et al., 2023), not other fully-engaged PIC structures.

Thanks for clarifying. We note that some structures of TFIIH alone also see the long helix. Accordingly, we modified this section to read:

“In many TFIIH and PIC structures the linker is not visible, presumably due to flexibility. However, when it is seen (Abril-Garrido et al., 2023; Greber et al., 2019), the linker emerges from the N-terminal domain as a long alpha-helix running along the interface between the two ATPase subunits…”

Page 8 Recent structures (reviewed in (Yu et al., 2023)) show that the Kinase Module would block interactions between the Core Module and other NER factors. Therefore, TFIIH either enters into the NER complex as free Core Module, or the Kinase Module must dissociate soon after.

To my knowledge, this is still controversial in the NER field. I note the potential function on the kinase module is likely attributed to the N-terminal region of Tfb3 through its binding to Rad3.

We are not experts on NER, but in reviews of the field this appears to be a widely held assumption. A 2008 paper from the Egly lab (Coin et al., DOI 10.1016/j.molcel.2008.04.024) is usually cited, which shows that the interaction between XPD (metazoan Rad3) and XPA is likely incompatible with XPD-MAT1 interaction. In addition to the Yu 2023 review, we now also cite a more recent publication that more extensively reviews the models for core TFIIH interactions (van Sluis et al, 2025). We looked at the multiple recently published structures of various TCR-NER and GG-NER intermediate complexes, and none of them show the CAK module or even the Tfb3/Mat1 N-term, even though those proteins were typically included during assembly. We also consulted with our colleagues Johannes Walter and Lucas Farnung, who are studying various TC-NER intermediates biochemically and structurally. Although the CAK module is included in their assembly reactions, it is not visible in their cryoEM structures. They tell me that the presence of CAK would be compatible with early TC-NER intermediates, but is predicted to overlap with later interactions of XPD with the TC-NER factor STK19 (see Mevissen et al., Cell 2024). To be conservative, we modified the sentence to say “Recent structures … suggest” rather than “show”.

Because the yeast strains used in Fig. 6 retain the N-terminal region of Tfb3, the UV sensitivity assay presented here is unlikely to directly address the contribution of the kinase module to NER.

We agree that our experiment only shows that the connection between Tfb3 N- and C-term domains is not necessary for NER. The individual domains might still be able to function independently. Accordingly, we changed the heading of that section from “Disconnected core TFIIH does not cause an NER defect” to “Split Tfb3 does not cause an NER defect.” This more closely matches the figure legend title.

Page 11. Notably, release of the Tfb3 Linker contact also results in the long alpha-helix becoming disordered (Abril-Garrido et al., 2023), which could allow the kinase access to a far larger radius of area. This flexibility could help the kinase reach both proximal and distal repeats within the CTD, which can theoretically extend quite far from the RNApII body.

Although the kinase module was resolved at low resolution in all PIC-Mediator structures, these structural studies consistently reveal the same overall positioning of the kinase module on Mediator, indicating that its localization is constrained rather than variable. This observation suggests that the linker region may help position the kinase module at this specific site, likely through direct interactions with the PIC or Mediator. This idea is further supported by numerous cross-links between the linker region and Mediator (Robinson et al., 2016).

That is true. But please note that this sentence was meant to describe movement of the kinase module AFTER release from Mediator (see previous sentence). Re-reading the passage, we realized the confusion is because we propose multiple possible pathways in that paragraph. In the first half, we suggest the capture of the kinase module by Mediator might trigger the conformation changes in the linker. In the second half (where it says “Alternatively….”) we suggest the Mediator-CAK interaction could instead come first, and the release of this contact could free the CAK module to move around. We have modified the paragraph to make it clear these are two different distinct models.

Comments on revisions:

Revised ms clarified all my points, including those I previously misunderstood.

Thanks again for helping us improve the manuscript.

Reviewer #2 (Public review):

Summary:

This work advances our understanding of how TFIIH coordinates DNA melting and CTD phosphorylation during transcription initiation. The finding that untethered kinase activity becomes "unfocused," phosphorylating the CTD at ser5 throughout the coding sequence rather than being promoter-restricted, suggests that the TFIIH Core-Kinase linkage not only targets the kinase to promoters but also constrains its activity in a spatial and temporal manner.

Strengths:

The experiments presented are straightforward and the model for coupling initiation and CTD phosphorylation and for evolution of these linked processes are interesting and novel. The results have important implications for the regulation of initiation and CTD phosphorylation.

Comments on revisions:

The revised version with revisions to figures, text and new data has addressed all of our prior comments.

We thank the reviewer for helping us improve the paper.

Reviewer #3 (Public review):

Summary:

Eukaryotic gene transcription requires a large assemblage of protein complexes that govern the molecular events required for RNA Polymerase II to produce mRNAs. One of these complexes, TFIIH, comprises two modules, one of which promotes DNA unwinding at promoters, while the other contains a kinase (Kin28 in yeast) that phosphorylates the repeated motif at the C-terminal domain (CTD) of the largest subunit of Pol II. Kin28 phosphorylation of Ser5 in the YSPTSPS motif of the CTD is normally highly localized at promoter regions, and marks the beginning of a cycle of phosphorylation events and accompanying protein association with the CTD during the transition from initiation to elongation.

The two modules of TFIIH are linked by Tfb3. Tfb3 consists of two globular regions, an N-terminal domain that contacts the Core module of TFIIH and a C-terminal domain that contacts the kinase module, connected by a linker. In this paper, Giordano et al. test the role of Tfb3 as a connector between the two modules of TFIIH in yeast. They show that while no or very slow growth occurs if only the C-terminal or N-terminal region of Tfb3 is present, near normal growth is observed when the two unlinked regions are expressed. Consistent with this result, the separate domains are shown to interact with the two distinct TFIIH modules. ChIP experiments show that the Core module of TFIIH maintains its localization at gene promoters when the Tfb3 domains are separated, while localization of the kinase module, and of Ser5 phosphorylation on the CTD of Pol II, is disrupted. Finally, the authors examine the effect of separating the Tfb3 domains on another function of TFIIH, namely nucleotide excision repair, and find little or no effect when only the N-terminal region of Tfb3 or the two unlinked domains are present.

Strengths:

Experiments involving expression of Tfb3 domains in yeast are well-controlled and the data regarding viability, interaction of the separate Tfb3 domains with TFIIH modules, genome-wide localization of the TFIIH modules and of phosphorylated Ser5 CTDs, and of effects on NER, are convincing. The experiments are consistent with current models of TFIIH structure and function and support a model in which Tfb3 tethers the kinase module of TFIIH close to initiation sites to prevent its promiscuous action on elongating Pol II.

We appreciate that the reviewer finds that our main conclusions are convincing.

Weaknesses:

The work is limited in scope and does not provide major insights into the mechanism of transcription. The main addition to current models of transcription is that tethering of Kin28 to Tfb3 may limit kinase action from occurring downstream from the initiation site.

The first described experiment, which purports to show that three kinases cannot function in place of Kin28 when tethered (by fusion) to Tfb3 is missing the crucial control of showing that Kin28 can support viability in the same context. This result also does not connect with the rest of the manuscript, although the experiment apparently motivated the subsequent studies reported here.

We elected not to do this control experiment for several reasons. As reviewer 3 points out, this kinase fusion experiment turned out to be somewhat disconnected from the rest of the paper. Even though it didn’t work, we included it in the paper because the results led us to the realization that the Tfb3 C-term was actually not fully essential for viability as reported, which in turn led us to the idea of splitting Tfb3. Structural studies (https://doi.org/10.1126/sciadv.abd4420, https://doi.org/10.1073/pnas.2009627117, https://doi.org/10.7554/eLife.44771) show that, in addition to providing linkage to the core module, the C-term of Tfb3 induces a conformation change in Kin28/Cdk7 necessary for full kinase activity (which is likely why the strains without C-term are just barely viable). If we were to pursue why the fusions didn’t work, we could tether Kin28 directly to the Tfb3 linker (and may try this in the future), but then would need to also express the C-term separately for its activating function. Even then, this would be an imperfect control for the fusion experiments in Figure 1. Because were trying to best mimic Kin28 being tethered via the accessory subunit Tfb3/Mat1, in the Figure 1 experiment we did not directly attach the kinases to Tfb3. For Ctk1/Cdk12, we fused the Tfb3 linker to the Ctk3 accessory subunit (analogous to Tfb3), and for Bur1/Cdk9, we fused to the cyclin subunit Bur2 (there is no known third subunit in this complex). The one exception was Mpk1, which has no partner subunits and is not a CDK. There are many reasons why this high-risk protein fusion experiment may not have worked, but chose not to pursue it further at this time.

Finally, the authors present the interesting and reasonable speculation that the TFIIH complex and connecting Tfb3 found in mammals and yeast may have evolved from an earlier state in which the two TFIIH subdomains were present as unconnected, distinct enzymes. It will be interesting to have this idea tested more thoroughly as more molecular evolutionary data becomes available.

Comments on revisions:

For the most part, the authors have satisfactorily addressed my previous critique. In particular, they have added to their discussion of evolutionary implications, and performed an experiment casting doubt on the assertion of a dominant negative effect, and as a consequence removed this claim from the manuscript. I also pointed out that the fusion experiments that lead off the Results section are missing the crucial control of including a Tfb3-Kin28 fusion. The authors have elected not to perform this control experiment, pointing out that even this control would be imperfect in some respects, and agreeing that this experiment is somewhat disconnected from the rest of the paper. The reason for including it, in spite of its somewhat tangential nature, is that it provides something of a rationale for the experiments that follow. I don't so much mind their retaining the experiment, as the absence of this control (and indeed, the results) does not so much impact the later results. However, I think if it is to be included, this shortcoming should be explicitly recognized, especially as a service to younger scientists who could benefit from an exposition that includes a thorough consideration of potential control experimenents.

We thank the reviewer for helping us improve the paper.

eLife Assessment

This manuscript reports a high-quality genome assembly of the European cuttlefish, Sepia officinalis, a representative species of the Cephalopod lineage. This solid work relies on current best practices in genome sequencing and assembly, combining PacBio HiFi long reads and Hi-C chromatin conformation capture, and on state-of-the-art comparative genomic analyses, including chromosome number evolution and analyses of expanded gene families. The resulting genome will be a valuable resource for researchers interested in cuttlefish biology and comparative genomics in general.

Reviewer #1 (Public review):

[Editors' note: this version has been assessed by the Reviewing Editor without further input from the original reviewers. The authors have carefully considered all the reviewers' comments. The newly added analyses, figures, and text sections are of high quality, and we commend the authors for their in-depth revision of the manuscript.]

This manuscript presents a high-quality, chromosome-level genome assembly of the European cuttlefish (Sepia officinalis), a representative species of the cephalopod lineage. Using state-of-the-art sequencing and scaffolding technologies -including PacBio HiFi long reads and Hi-C chromatin conformation capture - the authors deliver a genome assembly with exceptional contiguity and completeness, as evidenced by high BUSCO scores. This genome resource fills a significant gap in cephalopod genomics and offers a valuable foundation for studies in neurobiology, behavior, and evolutionary biology. However, there are several major aspects that need to be strengthened.

Reviewer #2 (Public review):

This paper concerns an interesting organism, Sepia officinalis. However, in the opinion of this reviewer, the paper reads somewhat like a genome report. The authors have used 23x PacBio HiFi in conjunction with relatively low coverage (11x) Hi-C to scaffold the genome into a karyotype of 47 chromosomes. They have used a combination of short and long read RNA seq to annotate the genome in what looks like a very good annotation. The paper offers basic analyses of the Busco evaluation, some descriptive analyses of gene family and repeat content, and a bit more focused analysis on synteny among sequenced squids. Generally, the data will be useful.

Reviewer #3 (Public review):

Summary:

In this study, authors Simone Rencken and co-authors present and investigate the genome of the common cuttlefish Sepia officinalis.

Strengths:

The authors explain in a detailed yet concise manner the main steps for a genome assembly, with very robust methods for validation, and according to current best practices. In addition to the chromosomal assembly, the authors confirmed the presence of 47 chromosomes using Hi-C data and multiple species synteny. They also generated a comprehensive gene annotation, with assessments of gene completeness, providing a useful resource for the community of researchers interested in cuttlefish biology and comparative genomics.

Author response:

The following is the authors’ response to the original reviews.

Public Reviews:

Reviewer 1 (Public review):

Summary:

This manuscript presents a high-quality, chromosome-level genome assembly of the European cuttlefish (Sepia officinalis), a representative species of the cephalopod lineage. Using state-of-the-art sequencing and scaffolding technologies -including PacBio HiFi long reads and Hi-C chromatin conformation capture - the authors deliver a genome assembly with exceptional contiguity and completeness, as evidenced by high BUSCO scores. This genome resource fills a significant gap in cephalopod genomics and offers a valuable foundation for studies in neurobiology, behavior, and evolutionary biology. However, there are several major aspects that need to be strengthened.

Major Revisions Recommended:

(1) Single-individual genome limitation

The genome assembly is based on a single individual, which appears to be male. While this approach is common in genome projects, it does not capture the full genetic diversity of the species. As S. officinalis exhibits a wide geographical range and possible population structure, future efforts (or discussion in this manuscript) should consider re-sequencing multiple individuals - of both sexes and from diverse geographic origins - to characterize population-level variation, sex-linked features, and structural polymorphisms.

We thank the reviewer for this summary and the important point raised. While sequencing additional individuals, unfortunately, lies outside the scope of our study, we used the published data from the DToL assembly (from a male individual from a different geographical origin) to begin to investigate their differences.

First, we attempted to create a mixed assembly from both datasets, as also suggested by Reviewer 2, to increase data coverage and genetic information. Even though the heterozygosity estimate is quite low (ca. 1%), the mixed assembly produced severely inflated and fragmented results, yielding an assembly ca. 3× larger than expected, with the top 46 contigs covering only ~5% of the total length - a sign of over duplication and failed haplotype collapse.

This result is not surprising when considering the assembly algorithms: most programs, including hifiasm used in this study, assume a single diploid individual (or a trio assembly including data from both parents), so using multiple individuals breaks this assumption. Assembly pipelines infer homozygous/heterozygous coverage cutoffs from the k-mer histogram. Mixing individuals raises apparent heterozygosity far above true diploid levels, turning the expected bimodal k-mer profile into a complex multimodal distribution. This misleads the phasing and purging steps in the assembly pipeline, causing over-expansion and fragmentation of the assembly.

Second, we created separate assemblies from the raw data sets of MPIBR and DToL using the exact same pipeline and parameters to avoid the technical problem described above. These assemblies are directly comparable, and after aligning them, it is possible to build a pangenome graph that we believe would help to address the points raised by the reviewer. Pangenome graphs can represent cross-individual variation more accurately and improve read alignment in regions of high genomic variation, which can aid population-level analyses [1]. We agree on the importance of this work, yet collecting data from more individuals and the construction and analysis of a pangenome graph lies beyond the scope of this manuscript and should be part of future efforts by the cephalopod genomics field.

(2) Limited experimental validation of chromosomal inferences

The study reports chromosome-scale scaffolding using Hi-C data and proposes a revised karyotype for S. officinalis. However, these inferences would be significantly strengthened by orthogonal validation methods. In particular, fluorescence in situ hybridization (FISH) or karyotyping from cytogenetic preparations would provide direct confirmation of chromosome number and structural arrangements. The reliance solely on Hi-C contact maps for inferring chromosomal organization should be acknowledged as a limitation or supplemented with such validations.

We appreciate the reviewer’s point regarding the value of orthogonal validation methods to support the chromosome-scale scaffolding and proposed karyotype. We acknowledge that relying solely on Hi-C contact maps to infer chromosome number and structure presents limitations, as also becomes apparent in our detailed analysis of both S. officinalis genome assemblies (in Figure 2 and Supplementary Figure 3 of the revised manuscript). We attempted to complement these analyses with cytogenetic approaches. Unfortunately, the availability of suitable mitotic tissue was limited. Moreover, our karyotyping trials proved challenging: resolving the ≥92 (2n) chromosomes in situ was not feasible due to their high number and the small size of the nuclei (approximately 5 µm in diameter on average).

We now highlight this point as an important direction for future work in our discussion (line 456-466):

“Additional methods such as cytogenetic karyotyping or optical mapping such as BioNano [141] (imaging of fluorescently tagged, linearized DNA) could be used to validate chromosome numbers. However, whereas karyotypes of octopuses have been consistent throughout the literature (1n=30) [142,143], those measured in decapods vary greatly. For example, 1n=46 chromosomes have been reported for two species of cuttlefish (A. esculentum and A. lycidas) and three loliginid squids [85]; 1n=36 has been reported for A. Arabica [86] and 1n=24 in A. pharaonis [87]. In S. officinalis, a karyotype of 1n=52 is reported for testis samples [88]. Combining cytogenetic preparations with fluorescent labeling of centromeric or telomeric sequences, as demonstrated in the octopus A. aerolatus [143] could help resolve these issues. Establishing a routine staining protocol would enable comprehensive tests at the species- and population-level.”

(3) Shallow discussion of chromosomal evolution

The manuscript briefly mentions chromosomal number differences among cephalopods but does not explore their evolutionary or functional implications. A more thorough comparative analysis - linking chromosomal rearrangements (e.g., fusions, fissions) with ecological adaptation, life history, or neural complexity - would greatly enhance the impact of the findings. Referencing chromosomal dynamics in related taxa and possible links to behavioral innovations would contextualize these results more effectively.

We agree with the reviewer that this is a fascinating topic of research that demands further attention and have extended our discussion, which now reads (line 476-501):

“In addition to studying chromosomal topology in phylogenetic reconstructions, some of the most interesting aspects of these rearrangements relate to changes of and innovation in regulatory elements that underlie phenotypic diversity. In coleoid cephalopods, it is thought that an ancient large-scale genome rearrangement was combined with lineage-specific changes and repeat expansions [48–50]. This restructuring gave rise to hundreds of tightly linked, evolutionarily unique microsyntenies, corresponding to distinct topological compartments with specialized regulatory architectures that contribute to complex, tissue-specific expression patterns in the nervous system and elsewhere [43]. Extending this, chromosomal conformation analyses in E. scolopes revealed that co-regulated eye and light-organ genes cluster at topologically associating domain (TAD) boundaries, and that an evolutionarily recent rearrangement at the dachshund (DAC) locus may have been instrumental in the emergence of the symbiotic light organ in Euprymna - directly linking specific chromosomal topology to morphological innovation [44].

To understand the broader functional impact of these changes across coleoids, a recent study investigating Micro-C, RNA-seq, and ATAC-seq data from multiple species revealed broadly conserved chromatin domains, but also many lineage-specific chromatin loops that form novel regulatory signatures and impact expression profiles across species and tissues [149].

Despite the observed small-scale regulatory changes, the chromosomes of decapods are considered to be more closely related to the ancestral coleoid karyotype than those of octopods. The derived octopod karyotype becomes apparent when comparing it to the genome of the vampire squid, an early-branching octopodiform (sister to all octopods) which retained features of the decapod, ancestral karyotype [150]. Taken together, the conserved karyotype of decapods accommodates fine-scale regulatory diversity that might underlie morphological diversity among species, which suggests that many regulatory innovations are still being evolutionarily explored through rearrangements within the existing chromosomes.”

(4) Underdeveloped gene family and pathway analysis

While the authors identify expansions in gene families such as protocadherins and C2H2 zinc finger transcription factors, the functional significance of these expansions remains speculative. The manuscript would benefit from:

(a) Functional enrichment analyses (e.g., GO, KEGG) targeting these gene families.

(b) Expression profiling across tissues or developmental stages to infer regulatory roles.

(c) Comparison with expression or expansion patterns in other cephalopods with known behavioral complexity (e.g., Octopus bimaculoides, Euprymna scolopes).

(d) Potential integration of transcriptomic or epigenomic data to support regulatory hypotheses.

We thank the reviewer for these constructive suggestions and have substantially expanded the functional characterization of expanded gene families in the revised manuscript.

To address points a) + b), we performed GO enrichment analyses for all expanded gene families (orthogroups), both for the largest gene families and the most significantly expanded families identified from our CAFE5 analysis. Further, we cross-referenced all S. officinalis members of each expanded orthogroup against differentially expressed genes in our bulk RNA-seq data from multiple tissues (initially collected to improve the gene modeling), allowing us to infer tissue-specific expression patterns for the expanded families.

To address point (c), the species-resolved copy-number profiles from our orthogroup analysis directly situate the S. officinalis expansions within the broader coleoid context, including O. bimaculoides, O. vulgaris, E. scolopes, and D. pealeii, enabling direct comparison of expansion scale and lineage specificity across species with varying degrees of behavioural complexity. We note that the C2H2 zinc finger and protocadherin expansions show distinct phylogenetic profiles consistent with independent radiations in octopods and decapodiforms, in agreement with recent studies.

Regarding point (d), no epigenomic data for S. officinalis was publicly available at the time of writing, thus we focused on the transcriptomic data from this study, as described above.

We describe this analysis in two additional results paragraphs to the manuscript, one modified (Figure 4) and two new figures (Figure 5 and Supplementary Figure 7), which are reproduced (lines 294-400):

“Analysis of expanded gene families

We sought to investigate the S. officinalis gene annotation and place it in the context of gene repertoires from other cephalopod or molluscan species. First, we collected available genome annotations from 12 other molluscan species (Table 2) and clustered them using OrthoFinder v.3.1.0 [122], resulting in 23,658 orthogroups, hereafter named gene families.

First, we investigated 36 of the gene families that contain more than 100 genes in any of the species, with 17 of these families containing at least one gene of S. officinalis, that reflect large-scale gene family expansions (Figure 4E). We used the InterProScan and eggNOG-mapper annotations to infer functional roles of these genes, selecting the most common gene annotation as the name of the gene family.

The zinc finger C2H2-type transcription factors (TFs) were grouped into three of the large gene families, with the largest family (OG0000000) only present in decapod cephalopods. This likely reflects the largely independent expansions in the octopod and decapod lineages that date back to a burst of transposon activity ca. 25 million years ago [46,48,49]. The largest expansion across mollusks occurs in the cadherin-like family (OG0000001): 310 in S. officinalis, 283 in D. pealeii, 209 in A. lycidas, 102 in O. vulgaris, 55 in O. bimaculoides, with low but non-zero counts in bivalves (C. virginica, M. gigas). This profile is consistent with the protocadherin expansion first described in O. bimaculoides [46] and subsequently shown to be present across cephalopods [48,49,123].

HPGDS (OG0000005, hematopoietic prostaglandin D synthase) is a glutathione-S-transferase family member that catalyzes the conversion of prostaglandins, which have well-described roles in immune responses in vertebrates and insects [124,125]. This family shows a broad expansion in decapods, with a lesser expansion in octopods. Additionally, members of the glutathione-S-transferase families have been co-opted as S-crystallins, structural proteins found in the lens of cephalopods that may, or may not, retain enzymatic functions [126,127].

Two large families are mostly lineage-restricted. The RING-type zinc finger family (OG0000058) has 103 copies in S. officinalis and 26 in A. lycidas but is absent in all other species except for E. scolopes. Conversely, OG0000002 (unknown function) has 479 copies in E. scolopes and only a few copies in the other species. This interesting Sepiolid-specific expansion warrants further characterization.

We estimated gene family evolution rates using CAFE5 [128] for all families with less than 100 copies in any species (this excludes the families described above, as very large copy-number differences between species preclude likelihood calculations under the applied birth-death model). After comparing different model parameters, we chose a gamma model with three rate categories, allowing for evolutionary rate variation among gene families. Out of the 12,895 gene families analyzed, 1,813 showed a significant (p < 0.05) expansion or contraction in at least one of the species. We focused our analysis on the 30 most significantly expanded families; among them were several retrotransposon-associated domains that have expanded specifically in S. officinalis five families carrying Retrovirus-related Pol polyprotein domains, two Reverse transcriptase domain families, and four Ribonuclease H-like families (Supplementary Figure 7A). There was no coordinate-based overlap of the coding sequences with annotated TEs from the RepeatMasker output (Methods).

In addition to the three large gene families of C2H2 zinc finger expansions, 45 gene families containing this TF type showed a significant change in the CAFE5 analysis. Notably, eight of the significant gene families, as well as four of the largest gene families, were annotated as CCHC-type zinc fingers, which contain a “zinc knuckle” motif that is characteristic of retroviral nucleocapsid proteins [129] and is functionally integrated in the genomes of several species, including humans [130].

Some gene families without any relationship to retrotransposons were also expanded. For example, the UGT2A1-related family is a UDP-glucuronosyltransferase, a class of enzymes central to phase II detoxification and conjugation of metabolites, reported in other mollusks in the context of environmental chemical tolerance [131], and in insects in the context of pigmentation [132]. We also detected a family of homeodomain-like proteins, representing an expansion of this important TF family.

Tissue-specific expression of expanded gene families

To place the identified gene families in a functional context, we profiled their expression in the bulk RNA-seq data (taken from multiple tissues of S. officinalis) used originally for gene modeling (Figure 5A). Principal component analysis (PCA) revealed the largest axis of variation in gene expression to separate brain tissues from peripheral tissues, with skin being the most transcriptomically distinct (Figure 5A), consistent with the high number of tissue-specific differentially expressed (DE) genes identified in non-neural tissues (Figure 5B). We identified the genes belonging to expanded families that were differentially expressed across tissues and enriched gene ontology [133,134] (GO) terms for them to gain additional insight. The large families excluded from CAFE5 modelling and the significantly expanded families identified by CAFE5 were analyzed separately.

Eleven of the largest gene families were expressed in our data (Figure 5C) and five had enriched GO terms (Figure 5D,E). Among them, the cadherin family showed brain-restricted expression and GO terms related to cell–cell adhesion and calcium binding, consistent with their role in neuronal connectivity and circuit formation [46,135]. Two C2H2 zinc finger gene families were expressed in the optic and vertical/subvertical lobes of the brain and in the skin, with GO terms related to DNA-binding, transcriptional regulation or development. The RING-type zinc finger family was expressed specifically in the skin, with GO terms including zinc binding and ubiquitin protein ligase activity, the canonical function of RING-domain E3 ligases [136]. Genes of the HPGDS/S-crystallin family were expressed in the brain (basal and optic lobes and posterior subesophageal mass) and skin, with GO terms related to glutathione metabolism, matching their described enzymatic function. We did not find expression in the retina, which is expected given that S-crystallins are expressed in lentigenic cells of the eye [42,137] and these cells were not included during sampling.

Among the 30 most significantly expanded families examined (out of 1,813 total), expression was widespread (20/30) and tissue-specific differential expression was common (17/30), suggesting that a substantial proportion of expanded paralogs represent functional coding sequences with specialized spatial deployment (Supplementary Figure 7B). Ten of the retrotransposon-associated families were differentially expressed in the brain (optic and vertical/subvertical lobes) and skin, arguing against these loci being inactive repeat fragments and supporting their inclusion as transcribed gene models. Two significantly expanded families showed both differential expression and enriched GO terms (Supplementary Figure 7C). The first was the UGT2A1-related family, which had the largest number of differentially expressed genes overall, with expression concentrated in the skin, retina and posterior subesophageal mass of the brain. Enriched GO terms matched the described enzymatic function for this family, namely UDP-glycosyltransferase activity. The second gene family was the homeodomain-like family with enrichment for DNA binding terms consistent with their role as transcription factors, and was preferentially expressed in the vertical and subvertical brain lobes with weaker expression in other areas.

Collectively, many differentially expressed genes from expanded families were restricted to specific tissues or brain subregions (Figure 5F and Supplementary Figure 7D), indicating that paralogs within an expanded family have adopted distinct spatial expression domains and possibly, specialized functions.”

Reviewer 2 (Public review):

Summary:

This paper concerns an interesting organism, Sepia officinalis. However, in the opinion of this reviewer, the paper reads somewhat like a genome report. The authors have used 23x PacBio HiFi in conjunction with relatively low coverage (11x) Hi-C to scaffold the genome into a karyotype of 47 chromosomes. They have used a combination of short and long read RNA seq to annotate the genome in what looks like a very good annotation. The paper offers basic analyses of the Busco evaluation, some descriptive analyses of gene family and repeat content, and a bit more focused analysis on synteny among sequenced squids. Generally, the data will be useful.

Strengths:

This is a high-quality annotation, and the data ultimately will be useful to other researchers. I appreciate trying to understand what's happening between assemblies of S. officinalis.

Weaknesses:

I don't believe the data at hand makes a strong case for the argument of 47 chromosomes. This is my biggest sticking point with the paper, and it is for a few reasons:

(1) The authors point to assembly differences between the DToL assembly and the one presented in the manuscript and seem to claim that DToL is incorrect. However, the DToL assembly (xcSepOffi3.1) is based on much deeper HiFi and HiC coverage than the one at hand (51x and 80+x respectively). There are many things to try here, including:

(a) Downloading the DToL data and reassembling using a common pipeline.

(b) Downsampling the DToL data to similar coverage as what the authors have achieved.

(c) Combining your data and that of DToL for even deeper coverage (heterozygosity is low enough that I don't imagine this impeding things too badly).

We thank the reviewer for these helpful suggestions and want to clarify that we did not seek to point out errors in the DToL assembly, but rather to investigate the unexpected discrepancies between the two assemblies. It is correct that the DToL data has a much higher coverage than our data. We followed the individual suggestions and incorporated them into the revised manuscript. We reproduce the relevant sections below, and provide additional information:

(a) Downloading the DToL data and reassembling using a common pipeline.

We downloaded the DToL data and reassembled it using a common pipeline, yielding the results listed in Author response table 1. The DToL assembly is more contiguous, which is mainly due to its higher HiFi coverage. It also receives slightly better BUSCO scores (computed using odb12 as recommended by Reviewer 3).

Author response table 1.

Full statistics of S. officinalis assemblies from two independent datasets, assembled using a common pipeline.

The updated manuscript now reads (lines 146-159):

“A chromosome-scale assembly for Sepia officinalis was released recently by the Wellcome Sanger Institute’s Darwin Tree of Life project [75] (DToL, GCA_964300435.1). That genome was assembled from a male individual using high coverage PacBio Sequel II (~51x) and Arima2 Hi-C (~80x) data, with a final assembly size of 5.8 Gb. The the haploid chromosome number was estimated to be 49. To compare both S. officinalis datasets directly, we downloaded the DToL data and created two new assemblies using the pipeline described above (hifiasm using PacBio HiFi and Hi-C data). The resulting assemblies were overall very similar, with the DToL assembly having a slightly higher contiguity (N50 length, see Table 1) and BUSCO completeness (Supplementary Figure 2A,B) due to their higher sequencing coverage.”

To further compare the two datasets, we added a new Figure 2 to the revised manuscript and the following paragraph to the results (lines 160-169):

“After scaffolding with YAHS, both datasets reached the previously identified chromosome numbers (1n=47 for MPIBR and 1n=49 for DToL, Figure 2A,B). To further investigate this surprising discrepancy, we aligned both assemblies using Winnowmap [89] to locate the differences between them (Figure 2C). We observed four “breakpoints” (BP) of chromosome scaffolds: one in the MPIBR assembly compared to DToL (BP1: DToL_5 = MPIBR_40+44) and three in the DToL assembly compared to MPIBR (BP2: DToL_31+40 = MPIBR_2, BP3: DToL_41+46 = MPIBR_6, BP4: DToL_44+45 = MPIBR_7). We also aligned the assemblies to the chromosome-scale genome of another cuttlefish Acanthosepion esculentum (1n=46, GCA_964036315.1). In this alignment, all four breakpoints were collinear with single A. esculentum chromosomes (Figure 2D).”

(b) Downsampling the DToL data to similar coverage as what the authors have achieved.

Instead of downsampling the DToL data, we decided to analyze the Hi-C and HiFi data for both assemblies, focusing on the four “breakpoints” between the assemblies and the A. esculentum genome that we described above. First, we performed a QC analysis of the Hi-C reads using pairtools [2], the result is visualized in Author response image 1. The percentage of valid Hi-C read pairs, i.e., cis pairs with insert distances of more than 1 kb and trans pairs, following the Dovetail genomics QC manual (https://dovetail-analysis.readthedocs.io/en/latest/whole_genome/qc.html). When Hi-C pairs were aligned to the primary contigs from hifiasm (as is used for scaffolding with YAHS), the DToL HiC data contains fewer valid read pairs (11.4%) than the MPIBR data (43.1%), possibly due to using a different tissue (eye vs. optic lobe) and HiC kit (Arima 2 vs. Dovetail OmniC) for the library preparation. Nonetheless, due to the much higher overall coverage, the amount of valid read pairs is still 2.35x higher for DToL (144,014,368 pairs) than for MPIBR (61,318,955 pairs). The higher trans fraction (i.e. HiC pairs across contigs) is dependent on the length of the primary contigs, so the higher trans fraction for the MPIBR data can be explained by the lower contiguity of its primary contigs. It is conceivable that for both assemblies, the low numbers of valid read pairs introduce a technical fragmentation of certain chromosomes, as indicated by the identified breakpoints (Figure 2).

Author response image 1.

Analysis of Hi-C read pairs from both S. officinalis assemblies. Hi-C reads were aligned to the primary contigs from hifiasm (as is used for scaffolding with YAHS) and analyzed using pairtools. Note the higher fraction of long-range contacts (at least 1 kb cis pairs or trans pairs) in the MPIBR data (top) compared to DToL (bottom). Due to overall higher coverage, the absolute number of read pairs is higher for DToL than for MPIBR data.

Second, we performed a detailed analysis of read coverage along the breakpoint junctions of the discrepant chromosomes/scaffolds between both assemblies. We included a description of the results and a new Supplementary Figure 3 in the manuscript, (lines 171-207):

“To better understand the potential cause of these divergent chromosome numbers, we analyzed the Hi-C and HiFi coverage in the breakpoint regions (Supplementary Figure 3A). First, we aligned the Hi-Fi reads to the scaffolds and extracted all alignments along the 200 kb terminal scaffold windows to find any notable drops in coverage, or reads spanning any of the scaffold junctions. We detected no spanning reads. This is not surprising given that no contigs were assembled at these sites, resulting in the observed scaffold junctions. More interestingly, we noted a ~5-fold decrease in HiFi coverage along the DToL scaffold_40 (part of BP2) relative to its flanking regions, indicating a highly repetitive, low-mappability region at this boundary.

Next, we realigned the Hi-C data to the scaffolded assemblies using bwa-mem2 [91] and extracted all trans HiC pairs (between-scaffold contacts) using pairtools [92]. We normalized trans HiC contacts to the scaffold length and compared contact rates between breakpoint scaffolds to the baseline contact rate (computed from pairs of scaffolds with a clear 1-to-1 match between assemblies), and the contact rate within scaffolds (intra-scaffold pairs) (Supplementary Figure 3B,C). The contact rates within breakpoints were consistently lower than within scaffolds, likely falling below the threshold to be merged during assembly. However, the contact rates at three of four breakpoints (BP1, BP3, BP4) were significantly elevated above the genome-wide background distribution (empirical p = 0.010, 0.005, 0.005 respectively), suggesting that they may represent intra-chromosomal contacts disrupted by a misassembly. Notably, BP2 was not significant (empirical p = 0.170), likely due to the low coverage and mappability around the DToL scaffold_40 boundary. Considered jointly, the three DToL breakpoint scaffold pairs showed significantly higher trans contact rates than the background (Wilcoxon rank-sum, one-tailed, U = 1771, p = 0.004).

Lastly, we analyzed the repeat landscape around the 200 kb scaffold ends using RepeatMasker [93] and the custom repeat library that we had generated for Sepia officinalis (described further below). Compared to control scaffolds of the same assembly, we observed consistently elevated repeat content at the breakpoint junctions (mean 71.5% vs 67.6% masked bases), with an enrichment of unclassified repeats (32.1% vs 30.0%), which could explain a repeat-driven assembly fragmentation or scaffolding failure. The BP2 DToL scaffold_40 junction window was 99.99% masked (99.2% unclassified repeats), providing a likely mechanistic explanation for both the HiFi coverage drop and the absence of a significant trans Hi-C signal at this breakpoint. Taken together, these analyses suggest that the different chromosome numbers across the two S. officinalis assemblies are due to technical reasons, caused by repeat-rich scaffold boundaries that impair HiFi and Hi-C read alignment and in turn, correct assembly in these regions.”

(c) Combining your data and that of DToL for even deeper coverage (heterozygosity is low enough that I don't imagine this impeding things too badly).

When combining the data to achieve a higher coverage, we ran into the assembly fragmentation issues detailed above in response 1) to Reviewer 1.

(2) Looking at Figure 1, there appears to be a misjoin at chromosome 42. Looking carefully at Figure S1, that misjoin does not appear on any of the panels - this is confusing. Given the size of that chromosome and the authors' chromosome numbering, I'm guessing this is a manual merge (as it's larger than most of the chromosomes numerically close (40, 41, 43, etc). Further, staring closely at Figure 1, there appear to be cross-scaffold contacts between 42 and 43 and 42 and 44. Secondarily there are contacts between 43 and 44. This bit of the assembly seems potentially problematic.

This is a great observation, indeed the HiC maps differ between Figure 1 and Figure S1. Figure 1 is the result of scaffolding with YAHS and manual curation, whereas Figure S1 was scaffolded using HapHiC. We updated the figure legend to clarify this important difference. HapHiC produces very clean contact maps without the need for manual curation, but when analyzed at a higher resolution, the tool broke many contigs and ultimately compromised the assembly quality, possibly due to our comparatively low HiC coverage. Thus, we preferred to use YAHS and manual curation, which is perhaps inherently error-prone, as becomes apparent in the regions of the assembly that are pointed out by the reviewer.

Reviewer 3 (Public review):

Summary:

In this study, authors Simone Rencken and co-authors present and investigate the genome of the common cuttlefish Sepia officinalis.

Strengths:

The authors explain in a detailed yet concise manner the main steps for a genome assembly, with very robust methods for validation, and according to current best practices. In addition to the chromosomal assembly, the authors confirmed the presence of 47 chromosomes using Hi-C data and multiple species synteny. They also generated a comprehensive gene annotation, with assessments of gene completeness, providing a useful resource for the community of researchers interested in cuttlefish biology and comparative genomics.

Weaknesses:

While the study touches upon the subjects of gene content, TE activity, or species-level comparisons, the study does not provide in-depth investigations of these.

We thank the reviewer for their positive assessment of our manuscript. We acknowledge the descriptive nature and limitations of our previous analyses of gene content, TE distribution, and species comparisons. Our focus for the initial submission was to provide a high-quality assembly that could serve as a resource for anyone interested in Sepia officinalis or related species. However, we agree that greater insight into genome content is valuable as well. In the revised manuscript, we included a more detailed analysis of expanded gene families and GO enrichment analysis of our bulkRNAseq data, which we summarized in response 4) to reviewer 1.

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors):

Minor Revisions Recommended:

(1) Figure and legend clarity

Several figures lack sufficient annotation. All figures, including supplementary ones, should include:

(a) Clear axis labels.

(b) Descriptions of statistical measures (n values, error bars, statistical tests).

(c) Legends that allow the figure to be understood independently of the main text.

We updated the figures accordingly.

(2) Terminology and formatting

(a) Consistency in gene and species nomenclature should be maintained throughout (e.g., italicizing gene names and Latin binomials).

(b) Ensure that abbreviations (e.g., Hi-C, BUSCO, FISH) are defined upon first use.

We updated the nomenclature throughout the text and checked the definition of abbreviations used in the text. Further, we updated the names of several cuttlefish species according to the recent revision of genera, e.g. Sepia esculenta was changed to Acanthosepion esculentum [3].

(3) Literature coverage

The references primarily focus on earlier studies from 2010-2020. It would strengthen the context to include recent high-impact studies on cephalopod genomics and chromosomal biology published in the last 3 years (e.g., 2022-2024).

We apologize for this oversight and have extended the manuscript to discuss more of these recent studies.

(4) Clarify methods

While the methods section is generally detailed, some critical aspects are underspecified:

(a) Parameters used in genome annotation tools (e.g., BRAKER, RepeatMasker).

We thank the reviewer for bringing our attention to this shortcoming, and have added the missing parameters to the methods section. Additionally, the full code is available at https://gitlab.mpcdf.mpg.de/mpibr/laur/cuttlefishomics/soffgenome

(b) Criteria for ortholog clustering and gene family expansion analysis.

The details have been added to the methods section, which now reads (lines 828-853):

“Orthogroups were inferred across 13 molluscan species (Table 2), including S. officinalis, using OrthoFinder v3.1.0 [122] with default parameters. The input proteomes included the longest protein isoform per gene for each species. The rooted species tree from OrthoFinder [182,184] was converted to an ultrametric tree using the R package ape [183] v5.8.1.

Gene families were filtered by removing orthogroups present in only a single species, and by separating orthogroups containing 100 or more gene copies in any species, as extreme copy-number differences in gene families prevent likelihood calculation under the applied birth-death model.

Gene family evolution rates were estimated using CAFE5 [128] v5.1.1 on the filtered orthogroups, using the ultrametric species tree as input. Four models were evaluated: the base model (single global lambda), and Gamma models with k = 2, 3, and 4 rate categories, which allow evolutionary rate variation among gene families. The Gamma k = 3 model was selected based on the best (lowest) final log-likelihood score. All subsequent statistical inferences were performed under this model.

For families showing statistically significant expansion or contraction (p < 0.05 after Bonferroni correction), branch-specific copy-number changes were extracted from the CAFE5 output. Families were categorized as S. officinalis-specific, coleoid-specific, or broad expansions based on the distribution of significant changes across the phylogeny.

To assess whether expanded gene families in S. officinalis contained genes derived from or embedded within repetitive elements, a coordinate-based overlap analysis was performed. For each gene in an expanded orthogroup, the overlap between its coding sequence (CDS) coordinates and RepeatMasker annotations was computed using bedtools intersect v2.30 [185]. To avoid double-counting when multiple repeat annotations overlapped the same coding bases, overlapping repeat intervals were merged per gene prior to summing covered bases, and the overlap fraction was computed as merged covered bases divided by total CDS length.”

(c) Thresholds or cutoffs for synteny or duplication detection.

We included the details in the updated methods (lines 755-781):

“Synteny analyses between all chromosomes of the compared species were performed using the R package GENESPACE v.1.2.3 [175] with default parameters, described briefly below. Protein sequence similarity was first estimated using DIAMOND2 [109] in fast mode, and orthogroups and pairwise orthologues were inferred using OrthoFinder v2.5 [176] with hierarchical orthogroups (HOGs) enabled. Prior to synteny inference, tandem arrays were condensed to their most central representative gene, and gene rank order was recalculated on these array-representative genes to reduce confounding effects of tandem duplication on collinearity detection.

Syntenic blocks were identified pairwise between all genome combinations using MCScanX [177], constrained to DIAMOND hits where both query and target genes belonged to the same orthogroup (onlyOgAnchors = TRUE). Initial anchor hits were clustered into large syntenic regions using a density-based spatial clustering approach (dbscan [178]), with a minimum block size of five anchor genes (blkSize = 5) and a maximum of five intervening non-anchor genes permitted within a block (nGaps = 5). Anchor clustering used a search radius of 25 gene-rank positions (blkRadius = 25). All hits falling within a syntenic buffer of 100 gene-rank positions around confirmed block anchors (synBuff = 100) were retained as syntenic. No secondary syntenic hits were included (nSecondaryHits = 0). Syntenic orthogroups were integrated across all pairwise comparisons and collapsed into a pan-genome annotation anchored to. S. officinalis was used as the reference genome.

Syntenic relationships were visualized as riparian plots and pairwise dotplots using the built-in plotting functions of GENESPACE v1.2.3. Riparian plots were constructed using physical chromosomal coordinates (useOrder = FALSE) with S. officinalis as the reference, displaying all three genomes. A second riparian plot was generated highlighting a region of interest. Pairwise dotplots were produced species for the S. officinalis–D. pealeii and S. officinalis–E. scolopes genome comparisons, displaying only synteny-validated hits (type = "syntenic") with a minimum synteny score of 10 (minScore = 10) and a minimum of 10 genes per chromosome pair required for display (minGenes2plot = 10).”

Reviewer #2 (Recommendations for the authors):

Line 153 should be supplemental Figure 3B.

The text was referring to the correct Figure 2B (three species synteny comparison). It is now updated to Figure 3B in the revised manuscript.

Reviewer #3 (Recommendations for the authors):

(1) L37: Perhaps add a comparison with other species (mammals, Drosophila, etc.) to put this number in context.

We agree with this recommendation and added numbers for Drosophila and mouse to the text (lines 40-45):

“Coleoid cephalopods (octopus, squid, cuttlefish) are a highly derived group of mollusks, characterized by the largest nervous systems among all invertebrates (ca. 500 million neurons in an adult octopus of which 200 million are in the central brain [1,2], compared to ca. 140,000 in the fruit fly [3] or 70 million in the mouse [4]) and specializations with a great historical importance for neuroscience (e.g., “giant axons” [5] and “giant synapses” [6–8]).”

(2) L51, 279: "Octopodiformes" is a superorder, not a genus or a species name. It should not go in italics.

We updated this throughout the text.

(3) L53: "even smaller" seems odd here, because the argument of the sentence is to stress the large genome size of Octopodiformes. Perhaps start the sentence by stating that it is sometimes smaller, but often larger.

We rephrased the sentence for clarity, it now reads (lines 55-58):

“While the genomes of Octopodiformes (Octopus, Eledone, Argonauta) are either smaller than (1.1 Gigabases or Gb [45]) or comparable in size to that of humans (around 3 Gb [46,47]) the typical genomes of Decapodiformes (squids and cuttlefish) often reach 6 Gb [48,49].”

(4) L90: What tool was used to estimate the k-mer distribution of the long reads? Jellyfish? FastK? It's not mentioned anywhere in the text.

(5) L95: What k-mer size did the authors use to estimate k-mer distribution?

We thank the reviewer for pointing out this missing information, and have included the details in the methods (lines 692-694):

“The k-mer distribution was estimated using Meryl [165] within the Merfin [166] package with a k-mer size of 21, and genomeGenome size was estimated using GenomeScope [77] from Illumina short reads and PacBio HiFi data.”

(6) L99: What about using the most recent BUSCO databases? odb12?

We thank the reviewer for this question, which prompted us to compute BUSCO scores using the more recent odb12 database. The results are shown in Supplementary Figure 2C. Both gene sets have been refined by including more species and using a more stringent filtering approach, so the more recent database contains fewer and more conserved genes [4]. For the mollusca gene sets, a great improvement in completeness was observed between odb10 and odb12 (Supplementary Figure 2C); the metazoan completeness was marginally increased. Therefore, we evaluated all new assemblies produced since the first submission with the odb12 database.

(7) L107: How many scaffolds were obtained in total? After manual curation, how many of the scaffolds were placed in the "correct" chromosomes? How many scaffolds were in the shrapnel? Were these scaffolds mostly repetitive regions? Or did they contain important genetic information?

These are important questions. To evaluate the content of the “shrapnel”, we split the manually curated assembly into the 47 chromosomes and the 1840 residual scaffolds, and computed BUSCO scores for both. While the 47 chromosome scaffolds contain the majority of conserved genes: C:92.9%[S:92.7%,D:0.1%],F:4.0%,M:3.1% with metazoa_odb12 and C:88.7%[S:88.0%,D:0.7%],F:4.4%,M:6.9% with mollusca_odb12, the unplaced scaffolds still contain a few BUSCOs: C:2.5%[S:2.4%,D:0.1%],F:2.4%,M:95.1% from metazoa_odb12 and C:1.9%[S:1.7%,D:0.2%],F:1.2%,M:96.9% from mollusca_odb12. Even if only a few BUSCOs are present on these scaffolds, it means they contain important genetic information. Additionally, we observed low, but non-zero alignment of RNA reads to these scaffolds. We observed a slightly elevated repeat content in the unplaced scaffolds (Author response image 2), and a variable base composition (Figure 1C) compared to the chromosome scaffolds.

Author response image 2.

Quantification of repeat content in chromosome scaffolds and unplaced residual scaffolds. Density plot showing fraction of repeat masked bases in total sequence length for chromosome scaffolds (i.e. scaffolds 1-47) in teal and all remaining small scaffolds (1840 scaffolds) in purple. Median repeat fraction is shown as vertical lines.

The slightly elevated repeat content in the unplaced scaffolds provides a likely explanation for their fragmented state: repeat-rich regions are inherently difficult to assemble and scaffold, as repetitive sequences cause ambiguous read alignments that prevent contigs from being confidently joined or anchored to chromosomal scaffolds during HiC-based scaffolding. This is consistent with the near-complete absence of BUSCO genes from the unplaced scaffolds - not because these fragments lack biologically relevant sequence entirely, as evidenced by the residual BUSCO hits and RNA read alignments, but because the gene-rich portions of the genome are largely captured in the 47 chromosome scaffolds. The unplaced scaffolds instead likely represent fragmented contigs from repetitive or low-complexity genomic regions, such as centromeres, telomeres, and transposable element clusters, where assembly graph complexity and collapsed repeats prevent confident placement. The variable base composition further supports this interpretation, as GC-extreme or low-complexity sequences are disproportionately represented in assembly shrapnel. Together, these observations suggest that the unplaced scaffolds contain limited unique coding content but reflect genuine repeat-rich genomic sequence that cannot currently be placed without additional long-range information, such as optical mapping or ultra-long reads.

(8) L33, 53, 240, 255, 279: Decapodiformes, not in italics.

We changed this throughout the text.

(9) L228: Can you put this expansion in perspective with other taxa?

We added a more detailed comparison of our gene family expansion with different species to the revised manuscript, as detailed in response 4 to reviewer 1.

(10) L251: "However, our results show how difficult it still is to assemble large genomes with high karyotype numbers." Can you clarify how your results show this, because it is equally spectacular to assemble the karyotype with only PacBio and Hi-C data (and no linkage mapping).

Indeed, it is correct that the recent improvements in data quality and scaffolding algorithms enable these “spectacular” chromosome-scale assemblies without the need for linkage mapping. This sentence reflected our expectation to resolve a clear karyotype as has been demonstrated for multiple cephalopod genomes in recent years, including two cuttlefish species (Octopus bimaculoides, Octopus vulgaris, Euprymna scolopes, Euprymna berryi, Acanthosepion lycidas and Acanthosepion esculenta). To our knowledge, none of these publications used linkage mapping or cytogenetic methods to confirm the karyotype. In this light, our resulting chromosome number and the discrepancy to a second assembly of the same species led us to this conclusion. We updated the section in the revised discussion as follows (lines 466-473):

“Taken together, our results illustrate the difficulty of assembling large genomes with high repeat content and large karyotypes, at least from sequencing data alone. Internal validation methods and genome comparisons across species are therefore important. Convergence of reliable estimates will, in turn, help identify chromosomal fusion-with-mixing events (FWM; fusion of two ancestral chromosomes followed by extensive shuffling of their gene content) that are clade specific. Early branching order in Decapodiformes has been notoriously unstable [53,84,94,144–147]; thus, such rare and irreversible FWM characters could be useful in further phylogenetic analysis of this clade [51,148].”

(11) L419: Why use the phased haplotype 1 instead of the primary assembly generated by hifiasm?

We thank the reviewer for this important question. We used the phased haplotype assembly because it provides a biologically coherent representation with the least amount of duplication by avoiding allele-collapsing and haplotype-switching that can be present in the primary assembly. We reasoned that this would result in clearer gene models and a more accurate representation of structural variation. However, we acknowledge that this comes at the cost of reduced contiguity and completeness, as becomes apparent in our BUSCO comparison shown in Supplementary Figure 2, where the phased haplotypes have fewer duplicated genes than the primary assembly, but more missing genes in turn. When reassembling both datasets for our comparison, we used the primary assembly to use the longest contigs as input for scaffolding.

(12) L444: It is unclear from what tissues and life stages RNA-seq data were used or were available from other species.

This is an important detail. RNA-seq data was collected from two adult Sepia officinalis, from various tissues (whole brain, retina, skin, mantle, arm, tentacle). For the long-read PacBio Isoseq data, tissue was taken from the animal used for genome sequencing (6 months old), and tissue for short-read Illumina RNA-seq was taken from another adult (8 months old). The data have been released on SRA (study accession SRP570862), where all sample details are listed as well. We added the SRA accession to the data availability section of the revised manuscript. We clarified the relevant sections in the methods:

lines 628-629:

“RNA was isolated from various flash-frozen tissues (different brain areas, mantle/epidermis, arm/tentacle; 5-10 mg each).”

lines 678-680:

“For short-read RNA sequencing, tissue from another animal (8-month-old adult, F0 from eggs collected in Normandie, France) was used. RNA was isolated from various flash-frozen tissues (different brain areas, skin and retina; 5 mg each).”

(13) L454, 469: Why is minimap2 in italics? It wasn't formatted like this before. Same for StringTie.

We thank the reviewer for their detailed methods review. In the updated methods section, all formatting of used softwares was harmonized.

(14) L461: Lophotrochozoa is a clade, not a genus or species. Not in italics.

This is now changed throughout the revised manuscript.

(15) Figure 1D: Axes labels are hard to read.

We have now increased the axis label size.

(16) Figure 2: Consider increasing font sizes. Many chromosome orientations seem to be flipped across species, which makes it harder to see smaller-scale rearrangements or notice less conserved chromosomes. Would it make sense to standardize these?

We increased the font sizes and plotted only fully collinear syntenic blocks (instead of aggregated syntenic regions, the default of GENESPACE) for improved readability.

References:

Below are references cited in our responses. References from the reproduced manuscript sections are included in the revised manuscript.

(1) Secomandi, S., Gallo, G.R., Rossi, R., Rodríguez Fernandes, C., Jarvis, E.D., Bonisoli-Alquati, A., Gianfranceschi, L., and Formenti, G. (2025). Pangenome graphs and their applications in biodiversity genomics. Nat. Genet. 57, 13–26. https://doi.org/10.1038/s41588-024-02029-6.

(2) Open2C, Abdennur, N., Fudenberg, G., Flyamer, I.M., Galitsyna, A.A., Goloborodko, A., Imakaev, M., and Venev, S.V. (2023). Pairtools: from sequencing data to chromosome contacts. Preprint at bioRxiv, https://doi.org/10.1101/2023.02.13.528389 https://doi.org/10.1101/2023.02.13.528389.

(3) Lupše, N., Reid, A., Taite, M., Kubodera, T., and Allcock, A.L. (2023). Cuttlefishes (Cephalopoda, Sepiidae): the bare bones—an hypothesis of relationships. Mar. Biol. 170, 93. https://doi.org/10.1007/s00227-023-04195-3.

(4) Tegenfeldt, F., Kuznetsov, D., Manni, M., Berkeley, M., Zdobnov, E.M., and Kriventseva, E.V. (2025). OrthoDB and BUSCO update: annotation of orthologs with wider sampling of genomes. Nucleic Acids Res. 53, D516–D522. https://doi.org/10.1093/nar/gkae987.

The paper has now been published under the title "Quantifying phage-host dynamics using droplet microfluidics" in Nature Communications:

Givelet, L., von Schönberg, S., Katzmeier, F. et al. Quantifying phage-host dynamics using droplet microfluidics. Nat Commun 17, 3857 (2026). https://doi.org/10.1038/s41467-026-72427-3