On 2024-10-29 10:08:15, user Jiayu Wang wrote:

Our database is available at https://datasync.ed.ac.uk/index.php/s/qlIgUlTTzW8KjXM

On 2024-10-29 09:45:14, user Stew Late wrote:

For Figures 4B and 4C, I suggest placing the "D. melanogaster" label in the same style as in Figures 4D and 4E.

On 2024-10-29 04:24:41, user Dan MacNulty wrote:

The age estimates are impressive. To help readers appreciate the significance of these estimates, consider adding more background about what’s currently known about the age of aspen clones in general. Do estimates exist for other clones? If so, how do they compare to your estimates? How does your aging method compare to previous methods?

On 2024-10-28 20:29:55, user Pedro H. Oliveira wrote:

Another recent large-scale screening of the bacterial defensome has been performed: https://www.nature.com/articles/s41467-024-46489-0

On 2024-10-28 15:02:24, user nuala oleary wrote:

Hi,

I wanted to point out a small issue with the references to NCBI Datasets. The author names are misformatted. It should read as follows:

O’Leary NA, Cox E, Holmes JB, Anderson WR, Falk R, Hem V, Tsuchiya MTN, Schuler GD, Zhang X, Torcivia J, Ketter A, Breen L, Cothran J, Bajwa H, Tinne J, Meric PA, Hlavina W, Schneider VA. Exploring and retrieving sequence and metadata for species across the tree of life with NCBI Datasets. Sci Data. 2024 Jul 5;11(1):732. doi: 10.1038/s41597-024-03571-y.

Thanks!

On 2024-10-28 09:39:36, user Isabella Capellini wrote:

A revised version of this manuscript is now available in Proceedings B:<br /> Mortlock E, Silovský V, Güldenpfennig J, Faltusová M, Olejarz A, Börger L, Ježek M, Jennings DJ, Capellini I. 2024 Sleep in the wild: the importance of individual effects and environmental conditions<br /> on sleep behaviour in wild boar. Proc. R. Soc. B 291: 20232115.<br /> https://doi.org/10.1098/rspb.2023.2115

On 2024-10-25 17:03:55, user Thomas Nicol wrote:

This article is now published in cardiovascular research https://doi.org/10.1093/cvr/cvad101

On 2024-10-25 12:37:11, user Yuta Takada wrote:

There was a mistake in line 80. The correct reference is figure S1A, not figure S2A.

On 2024-10-24 17:27:06, user Kyle Barrie wrote:

Published: https://www.nature.com/articles/s41594-024-01412-5

On 2024-10-24 16:51:29, user Allisandra Rha wrote:

Interesting work. The multimodal approach definitely enhanced the integration outcomes and is well-designed. While it is of benefit that AAV efficiently reaches the nucleus, moving forward would you consider evaluating other delivery methods for nuclear translocation to account for patients that are seropositive for AAV antibodies?

Also, the paper is under-referenced. I would encourage you to look at the text and provide more citations to support your claims.

On 2024-10-23 00:20:34, user CDSL JHSPH wrote:

I really enjoyed reading your preprint "Universal rules govern plasmid copy number." Your analysis of plasmid copy number across diverse bacterial genera is impressive and the discovery of a scaling law linking plasmid size and PCN is a huge contribution to the understanding of plasmid biology. I was wondering, based on your discussion on the intrinsic variability of PCN in high-copy number plasmids versus low-copy number plasmids, how these variations impact the fitness of bacterial populations, especially under different selective pressures. Additionally, given plasmids' role in gene transfer, how might the scaling law and PCN-size trade-off influence the evolution of plasmid-encoded traits, such as antibiotic resistance genes or virulence factors? <br /> I look forward to seeing how this research progresses! Thank you!

On 2024-10-19 00:01:15, user CDSL JHSPH wrote:

Thank you for sharing your research! I found your paper on PCN very insightful, especially the discovery of "universal rules" governing PCN. The large-scale analysis of plasmids provides valuable insights into plasmid number variability and regulation, with potential applications in both research and biotechnology. I do have a few questions. Could replicon dominance bring evolutionary advantage, particularly under natural conditions where plasmids may compete for replication resource? Additionally, under external constraints, is there a threshold where dominance shifts between replicons? <br /> I very much enjoyed the article, and look forward to how these findings evolve in future studies.

On 2024-10-23 00:19:08, user CDSL JHSPH wrote:

I really enjoyed learning about this new Bento Lab because it opens the doors for so many new positive things. I think the paper is written in a way that people will understand the differences between BL and TL and agree that there are so many advantages to the BL. I think your strongest part is how you explain your data from the figures since you talk about the how and identify how to understand the figures and tables. I would say for figure one; I was wondering if maybe also using bar graphs since it shows a clearer way of differentiation like Figure 2. I also think instead of tables, more figures will help the public understand their pros and cons better visually. I also think that the sludge's low read count should be addressed more. While it is mentioned that the Flongle's limitations might have affected the detection of less abundant entities, why do you think happened to that sample, and how could you express that to the audience? I think you could strengthen their argument by referencing specific species, ARGs, or plasmid types they suspect might have been missed due to the lower read count. To end this, I think the authors did a great job explaining the differences, and my only concerns were adding figures and explaining the sludge sample.

On 2024-10-19 18:05:46, user CDSL wrote:

This article is a good demonstration of the potential of portable DNA sequencing technology for rapid detection of pathogens and antimicrobial resistance in the field, especially for public health emergencies in low-resource or remote areas, and the results section, in particular, is very detailed. However, I feel that there are some deficiencies in the discussion part. It is mentioned in both the introduction and the results that the differences are evaluated from five aspects, but I do not seem to see obvious discussion about DNA yield and purity in the discussion part. Also, have you considered separating the restrictions into a separate section? That may help the reader to read in a more organized way.

On 2024-10-22 20:43:09, user Thomas Sorger wrote:

Erratum: The tree in Panel A of Figure 9 is incorrect. <br /> It will be corrected in a forthcoming revision.

On 2024-09-10 18:32:52, user Thomas Sorger wrote:

Please note that the date of the second reference (2003) is incorrect. The correct date is 1983:<br /> 2. Armstrong E. Relative brain size and metabolism in mammals. Science 220: 1302–1304 (1983).<br /> Tom Sorger

On 2024-10-22 19:27:34, user Andy Flies wrote:

This chapter is now published.<br /> Slyp, B., Darby, J.M., Flies, A.S. (2024). Conversion of Mouse-Derived Hybridomas to Tasmanian Devil Recombinant IgG Antibodies. In: Good-Jacobson, K.L. (eds) Memory B-Cells. Methods in Molecular Biology, vol 2826. Humana, New York, NY. https://doi.org/10.1007/978-1-0716-3950-4_17

On 2024-10-22 19:01:27, user CDSL JHSPH wrote:

I believe this is an excellent study and paper, with valuable contributions to the field. It was very well-written, and the methodology very clearly described and reproducible. Given the global clinical significance of malaria, the biological implications of this study can be used as a basis for further studies into the biology of circadian rhythms in malaria transmission, as well as clinical translation for efficient prevention and control strategies and more effective treatments.

Similar to what previous reviewers have noted, I noticed some inconsistencies between some of the figures and the apparently corresponding results in the text, especially as it relates to the percentages of the transcriptome displaying cyclic expression profiles, for both the mosquito salivary glands and sporozoites. The figures and/or results should therefore be corrected accordingly.

A possible limitation to the study, which I don't believe was addressed in the paper, was that different mosquito (Anopheles stephensi vs Anopheles gambiae) and parasite (Plasmodium berghei versus Plasmodium falciparum, vivax, ovale, malariae, or knowlesi) species from the ones which are primarily known to cause malaria in humans, were used for analysis in this study. This may limit direct clinical translation of these results for malaria in humans. There should be a statement addressing this in the discussion. Furthermore, the amount of gene homology between the species used in the study and the species causing disease in humans should be stated for reference. If there were any reasons it was not possible or necessary to use the species known to be most associated with disease in humans, this should be stated.

On 2024-10-22 01:49:09, user CDSL JHSPH wrote:

I thoroughly enjoyed reading your paper, and I believe this study brings valuable new insights into how circadian rhythms influence malaria transmission. This information will undoubtedly impact future strategies for combating malaria, but also other vector-borne diseases. However, I noticed some inconsistencies between the results and the corresponding figures. I might be wrong and misread it but I wanted to bring it to your attention and clarify it on my side as well. In certain sections, figures were cited but did not seem to align with the content discussed. I can mention for example:<br /> - “We found that 27-49% of salivary gland genes displayed a cyclic expression profile, (Fig. 1B-E) representing 5-10x more than what has been previously reported for the mosquito’s head or body”. For C-E, I don’t see how this is actually depicted. Could you please explain; even the next sentence as well I don’t understand how C-E are supposed to back the arguments you are trying to make <br /> - Some figures like 1G and 1H are mentioned in the legend of your figures and in the text as well but are not shown in the figures, why?<br /> Those are not the only inconsistencies I noticed, but before proceeding on citing them, I want to make sure that I am not misreading and not understanding.

If my assumptions are true, clarifying them would greatly improve the paper’s clarity and presentation of your significant findings. I am looking forward to reading your answer.

Best regards,<br /> Critical Dissection Class 2024/2025

On 2024-10-19 02:11:16, user Yak Nak wrote:

The manuscript provides an exciting and valuable look into how circadian rhythms influence malaria transmission by aligning mosquito feeding behavior and parasite activity. The use of RNA-sequencing to uncover rhythmic gene expression in mosquito salivary glands is a significant strength and offers important new insights into the mechanisms behind malaria transmission. The figures are clear and effectively illustrate how these rhythms correlate with mosquito feeding efficiency and parasite infection capabilities, though the figure legends could benefit from more detailed explanations, especially for readers unfamiliar with gene expression data. The introduction is solid but could be improved by providing a more detailed discussion of previous research on circadian rhythms in malaria parasites to better frame the novelty of this study. The discussion section does a good job connecting the findings to broader vector-borne diseases like Zika and dengue, but it would be even stronger with specific examples of how these results could inform practical malaria control strategies, such as optimizing the timing of interventions based on mosquito feeding times. Overall, this is a well-conducted study with important findings, and a few revisions could further enhance its clarity and impact. Two follow-up questions:

1. Could the authors clarify why specific time intervals (every 4 hours) were chosen for RNA-sequencing, and would more frequent sampling provide additional insights?
2. Also, how might environmental factors such as temperature or humidity influence these circadian rhythms, and could this affect transmission in different regions?

On 2024-10-18 18:57:00, user CDSL JHSPH wrote:

I really like this article. Thank you for sharing your article. Malaria remains a major global health challenge, and understanding the biological rhythms of mosquito vectors and malarial parasites is crucial for improving control strategies. However, previous studies have only revealed that mosquitoes have daily rhythmic behaviors. You provided new insights. First, about half of the genes in the mosquito salivary gland transcriptome show 24-hour rhythmic expression, and second, the gene expression of sporozoites in the salivary glands also shows circadian rhythms (parasite movement and infection ability). In addition, you mentioned in the discussion that the parasites and mosquitoes have evolved in coordination with the circadian rhythms, and jointly affect host infection by regulating parasite movement and mosquito blood feeding time. This is very interesting, and your findings provide a new perspective for understanding the temporal regulation mechanism of malaria transmission. At the same time, you mentioned optimizing the insecticide spraying time strategy according to the peak period of mosquito activity. In addition, your findings may also have reference effects on other diseases.

But I have some questions. First, it is good that you use 12h dark and 12h light to simulate day and night. But in the wild, the temperature and humidity vary between day and night. Perhaps the natural day-night cycle could be better simulated by changing temperature and humidity. Secondly, you mentioned that some observations may apply to uninfected mosquitoes, but you did not specifically discuss the potential differences in rhythms and behaviors between uninfected and infected mosquitoes.

Finally, thank you again for your contribution and new ideas.

On 2024-10-21 14:16:12, user MJ Muzzatti wrote:

This article is now published in the Journal of Economic Entomology http://dx.doi.org/10.1093/jee/toae208

On 2024-10-21 14:15:22, user Julian C wrote:

Accepted for publication in Journal of Proteome Research (comment added by the corresponding author).

On 2024-10-20 08:57:36, user Alireza Mani wrote:

This paper is now published at https://doi.org/10.1016/j.bbrc.2024.150854

On 2024-10-20 07:36:15, user XKM wrote:

This study has now been published in PLoS Biology.<br /> https://doi.org/10.1371/journal.pbio.3002615

On 2024-10-19 19:13:36, user sam wrote:

It would be good to see how altered the phosphoproteome is after FACS. Thats a long time for the cells and a lot can change in phosphopatterns in a matter of seconds .

On 2024-10-19 18:51:50, user Alessio Ciulli wrote:

This comment is to advise readers that our preprint has been published in Sci. Adv. 10, eado6492 (2024). https://doi.org/10.1126/sciadv.ado6492

On 2024-10-19 18:49:33, user Alessio Ciulli wrote:

This comment is to advise readers that our preprint has been published in Nat Commun 15, 8885 (2024). https://doi.org/10.1038/s41467-024-52871-9

On 2024-10-19 00:19:59, user CDSL JHSPH wrote:

This article is rigorous. I like the comparison between phages in gut and that in the sea water. And the analysis of phages in hCom2 mice is well designed. Yet I think the equations and the explanations in the result and the method part are not very readable. I’m confused about the meaning of lysogen-host ratio and relative coverage, the role of dilution rate and so on. Plus, figure 2B is not very informative and by just looking at it, I cannot have a clue of what you were doing to get the range of induction rate. But overall your results support your conclusions firmly and clearly.

On 2024-10-18 13:39:11, user Oscar wrote:

Dear James and co-authors,

I trust you are already working on submitting this manuscript to a reputable journal. As one of the researchers with the most publications on CAV1 in Ewing Sarcoma (Oscar M. Tirado), I would like to offer some comments and suggestions for your consideration:

It might be beneficial to mention that CAV1 is a demonstrated direct target of EWS::FLI1, as this could add further context to your findings.

Regarding the presence of caveolae in Ewing Sarcoma cells, I assume you've reviewed the manuscript (DOI: 10.1016/j.canlet.2016.11.020). In most of the cells studied, including TC71, the levels of Cavin-1, as well as caveolae, are extremely low. As such, the majority of CAV1’s functions in EwS are likely independent of caveolae, which aligns with the results you’ve presented.

I noticed the focus on the AKT pathway in your work. Have you considered discussing the MAPK pathway as well? It is also affected and likely plays a significant role in the processes you're describing.

In Figure 4, CAV1 emerges as a potential driver, along with EphA2. I’m sure you are familiar with this study as well (DOI: 10.1002/ijc.31405), which may provide further insights.

I agree with your assessment that CAV1 plays a key role in the progression of EwS, and your findings align with many of the conclusions we've drawn in our own research. I would be happy to assist further, either through this platform or via personal contact, as we've known each other for many years from various EwS meetings.

Best regards,<br /> Oscar M. Tirado

On 2024-10-18 01:51:53, user Tushar R. wrote:

Summary

Here, the authors sought to apply cryo-EM guided metainterference-based MD simulations to find modeling inaccuracies of flexible helical regions linked to their artificial representation by a single-structure model. They first applied their approach to the group II intron ribozyme from Thermosynechococcus elongatus, one of a few cryo-EM structures available in the PDB that featured mostly RNA and was obtained using a single-structure refinement procedure. They confirmed that remodeling with their approach mostly affected flexible regions at the solvent-exposed stem loops that were not phylogenetically conserved. They found that functional domains that were well-ordered, phylogenetically conserved, and were clearly represented by the cryo-EM density required no remodeling. Extending this analysis through other PDB structures revealed that poor modeling at flexible helical regions was broadly applicable to all RNA-containing cryo-EM-derived structures in the 2.5 - 4 Å resolution range.

Major Points:

It is difficult to keep track of the different parameters applied for each simulation. These are scattered in the text. It would be very helpful for the reader to understand these if the authors could make a table of all the MD simulations with specifications on helices restrained, approach, simulation time, force fields present, equilibration time etc.

Clarification regarding the helical restraints and the simulation time for the initial production run would be helpful—i.e. Why was a simulation time of 2.5 ns chosen, and would this be long enough for your purposes considering the relatively high complexity of the system?

Were restraints applied to the 3 helices (b,d,i) that unfolded in subsequent simulations or were they only applied to the other 6 helices?

In the section “Base Pairing Analysis of the Protein Data Bank,” it is mentioned that the analysis likely over-estimates the problematic modeling of helices in cryo-EM derived RNA structures. Given that 15 Å is larger than the expected 8-11 Å (Pietal, M. J., 2012) distance for N1/N9 distance, would this not underestimate the number of problematic helices?

A detailed analysis of H-bonding within the 6 helices would be useful to get a mechanistic understanding of why the 3 helices (b,d,i) unfold and the other 6 don't—for example, it might be helpful to provide a comparison with the model generated from a single structure approach to know if the same H-bonds exist in it or not as the model forwarded by this paper.

ERMSD approach: we know relatively less about non-Watson Crick base pairing apart from Hoogsteen and G-U wobble base pairs since so much depends on the context (specific region of the structure, ion concentration, pH etc.) (we’re still thinking about how to phrase this point).<br /> Balance between experimental measurements and MD sampled conformational states: lack of experimental validation when remodeling mismodeled regions for RNA - how do we know when specific interactions are too ideal?

How is variability between forcefield parameters for divalent metal ions or other ions addressed, especially considering that these are required for proper folding of RNA?

Dispersive interactions play a huge role in RNA integrity: how confident can we be about these parameters when comparing various nucleic acid specific force fields? Can these differences lead to unfolding of the 3 properly base paired helices?

With reference to the line on page 4: “The trajectories obtained with metainference simulations were then analyzed by back-calculating the corresponding averaged density map and comparing it to the experimental one”, was the solvent density included in the back-calculated density?

What proportion of the CC_mask arises just due to fitting of rigid part of the structure and how much of it arises due to the improved fitting of the flexible regions from the 9 loops? Would it be possible to separate these contributions?

Minor Points:

Would it be possible to show Fig 2C as a violin or box plot rather than a bar plot? This would allow visualization of the distribution of CC_mask for different conditions with clusters representing conformational states that may agree with the experimental data.

Is local resolution considered while plotting the RMSF? For this purpose, it would be useful to have a local resolution estimation for the map to help the reader understand whether the unfolding of helices occurs simply because the specific regions were not well defined or if these regions are inherently flexible.

How effective is DeepFoldRNA in filling gaps in structure as it is normally known for sequence based structure prediction—specifically regarding the modeling of the 38-nucleotide gap?

For benchmarking purposes, applying this approach to other RNA structures or providing additional validation against independent experimental data (e.g., SAXS or chemical shifts from short RNA motifs) could further strengthen the conclusions.

Aside from the pre-1r (6ME0) and pre-2r (6MEC) states, Haack et al. (2019) found additional 3D classes that indicated disordered density in conserved regions, which they did not investigate further. They also suggested that the post-2r complex may be captured in one of the 3D classes that yielded a low-resolution 3D reconstruction. Is it possible that one of the structures found by the metainference method could have sampled one of the structures that were not investigated further or the post-2r complex?

Why is there a sudden fall in the number of helices formed in 32 replicas (Fig 2E)?

Sentence on page 4: “As a result, 32 replicas appeared to be the best compromise between agreement with experiment and computational cost.” Has it been ensured that this is not a result of overfitting? One of the ways to do this can be to use half-map cross validation i.e. to check if the refined ensemble fits both the half maps equally.

Have the authors tried to change the weight for 1 μs reference simulation to improve the CC_mask?

Grammar/spelling mistakes <br /> Results, Test System and Preparation, 1st Paragraph: Change “run” to “ran” to retain past test throughout the paragraph in the sentence beginning “We then run 2.5 ns-long molecular dynamics (MD) simulations in explicit solvent…”<br /> Results, Ensemble Refinement, 1st Paragraph: “Within” is misspelled as “whitin” in the sentence beginning “Conversely, when restraining their helicity, whitin this single-replica refinement approach…”<br /> Results, Ensemble Refinement, 1st Paragraph: “Specifically, all the simulations attempted with a lower number of replicas were crashing reporting missing convergence in enforcing bond constraints, which indicates that the experimental and helical restraints were mutually incompatible.”

• Tushar Raskar, Sonya Lee, James Fraser

On 2024-10-17 18:28:16, user Christine Pye wrote:

This article has now been published in the Journal of Small Animal Practice https://doi.org/10.1111/jsap.13688

On 2024-10-17 14:50:43, user Elena Grassi wrote:

The revised version is now out:<br /> https://www.nature.com/articles/s41467-024-51909-2 <br /> https://doi.org/10.1038/s41467-024-51909-2

On 2024-10-17 14:12:50, user Sahil Sahni wrote:

Article published: https://www.nature.com/articles/s41467-024-52555-4

On 2024-10-17 14:05:57, user Christina Warinner wrote:

It is very nice to see the authors statistically confirm on a large number of samples that oral bacteria contribute to the thanatomicrobiome of archaeological teeth. They may want to note that this pattern has been previously observed and reported twice before: <br /> Mann AE, Sabin S, Ziesemer KA, Vågene Å, Schroeder H, Ozga A, Sankaranarayanan K, Hofman CA, Fellows-Yates J, Salazar Garcia D, Frohlich B, Aldenderfer M, Hoogland M, Read C, Krause J, Hofman C, Bos K, Warinner C. (2018) Differential preservation of endogenous human and microbial DNA in dental calculus and dentin. Scientific Reports 8:9822. DOI: 10.1038/s41598-018-28091-9<br /> Vågene AJ, Campana MG, Robles García N, Warinner C, Spyrou MA, Andrades Valtueña A, Huson D, Tuross N, Herbig A, Bos KI, Krause J. (2018) Salmonella enterica genomes recovered from victims of a major 16th century epidemic in Mexico. Nature Ecology and Evolution 1-9. DOI:10.1038/s41559-017-0446-6.

On 2024-10-16 00:18:50, user Hamidreza Karbalaeiheidari wrote:

Hello,<br /> My preprint has now been published in ACS Synthetic Biology Journal.<br /> https://doi.org/10.1021/acssynbio.4c00437

Thanks,<br /> Hamid Reza Karbalaei-Heidari

On 2024-10-15 20:51:50, user Alexa Jennings wrote:

Overall, this paper is significant to developmental biology. Thank you for your contribution and dedication to the field. As I read through your paper, I have a few comments I would like to offer.

Fig1: While I was intrigued by the several stainings and experiments carried out by your team, it would be helpful to understand your rationale for choosing the gestational ages of interest. Additionally, since this is a comparative review, it would be useful to visualize the corresponding gestational ages in mice. I also noticed you cited these phases in mice, but it would be much more memorable if there was a clear schematic comparing the two species' gestational ages to go along with the rest of the figure as well.

Fig2: I noticed a one-way ANOVA was carried out for several of the corresponding data. However, ANOVA's are designed to compare means of groups with samples >3. In the supplemental figures, several groups were either missing data, or had samples <3. Asterisks marking significance would also be useful to visualize significance on the graphs.

Fig3&4: Quantification of fluorescence would be useful to make statistical comparisons. As is, the evidence is too correlative to make any conclusions.

Fig5: For clarity, place comparative groups in alphabetical order (ex: 5a and 5c should be 5a and 5b). Additionally, 5k-q show dim fluorescence, and should be quantified to make accurate statistical comparisons. 5f should be compared via a multivariate-ANOVA.

Final remarks: Further manipulative, molecular experiments are required to make conclusions on similarity. A stronger argument could be made if additional evidence were included. Additionally, sex and cultural differences likely will cause variation among your data. Stating the sex of the samples can help make more accurate comparisons between data. Comparative experiments between sexes and across ethnicities in humans may also be useful.

On 2024-10-15 19:33:41, user Abdul Muntakim Rafi wrote:

the paper is now published at Nature Biotechnology<br /> https://www.nature.com/articles/s41587-024-02414-w

On 2024-10-15 18:37:28, user Jacek Majewski wrote:

This article has been published in Cell Reports:

Chromatin dysregulation associated with NSD1 mutation in head and neck squamous cell carcinoma

Cell Rep. 2021 Feb 23;34(8):108769. doi: 10.1016/j.celrep.2021.108769

On 2024-10-15 07:39:58, user turnersd wrote:

The paper states:

The source code is publicly available at https://github.com/ginnm/PROEDIT .

But the provided repository https://github.com/ginnm/PROEDIT is empty (README and license files only).

On 2024-10-14 20:52:21, user Patrick Varga-Weisz wrote:

We made basically the same observation in 2018 with crotonate and published in DOI: 10.1038/s41467-017-02651-5 | http://www.nature.com/naturecommunications <br /> (Fellows et al, Nature Coms 2018) see Figure 7d.<br /> Kind regards,<br /> Patrick Varga-Weisz

On 2024-10-14 06:17:39, user Ian Ross wrote:

As of Aug 23 (2024) the published version of this article is available at Plant Physiology Advance Articles https://academic.oup.com/plphys/advance-article/doi/10.1093/plphys/kiae427/7740471

On 2024-10-14 05:52:34, user PanosMoschou wrote:

Information missing from the Fig. legend of Fig. S1 panel I, in this version (data not present in the published version of the article in Plos Biology) represents a reconstructed image that fully reproduces the original dataset (in particular, the mating controls DDO and QDO experiments). If you wish to have access to the original dataset, don't hesitate to get in touch with the corresponding author. Furthermore, the group will soon post the original dataset in Zenodo.

On 2024-10-11 22:57:08, user aquape wrote:

Recent information, google<br /> "GondwanaTalks Verhaegen English"<br /> + see <br /> - M.Vaneechoutte cs 2024 Nat.Anthrop.2,10007 open access https://www.sciepublish.com/article/pii/94 <br /> "Have we been barking up the wrong ancestral tree? Australopithecines are probably not our ancestors” <br /> - id.2024 Nat.Anthrop.2,10008 https://www.sciepublish.com/article/pii/187 <br /> "Reply to Sarmiento E. “Australopithecine Taxonomy and Phylogeny and the Savanna Hypothesis…”"<br /> https://www.gondwanatalks.com/l/the-waterside-hypothesis-wading-led-to-upright-walking-in-early-humans/

On 2024-10-11 20:15:33, user Velizar Shivarov wrote:

May be you can refer to a recent publication: https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2024.1427810/full

On 2024-10-11 09:50:17, user bao zhang wrote:

I tried to run the script you uploaded on github, and the model.py script reported an error, "NameError: name 'p_SdC' is not defined", how is this p_SdC defined? I am looking forward to your reply, thank you.

On 2024-10-11 08:52:29, user Esther Broset wrote:

This article is highly informative, and the ionizable lipids show great promise. I was particularly impressed with the targeting strategy and the high expression levels observed in the lungs.

On 2024-10-11 08:50:00, user George Brown wrote:

Link to published article<br /> https://link.springer.com/protocol/10.1007/978-1-0716-3906-1_1

On 2024-10-10 22:17:31, user Delphine Destoumieux-Garzon wrote:

Now published in Science Advances:<br /> Gawra J, Valdivieso A, Roux F, Laporte M, de Lorgeril J, Gueguen Y, Saccas M, Escoubas JM, Montagnani C, Destoumieux-Garzόn D, Lagarde F, Leroy MA, Haffner P, Petton B, Cosseau C, Morga B, Dégremont L, Mitta G, Grunau C, Vidal-Dupiol J. Epigenetic variations are more substantial than genetic variations in rapid adaptation of oyster to Pacific oyster mortality syndrome. Sci Adv. 2023 Sep 8;9(36):eadh8990. doi: 10.1126/sciadv.adh8990.

On 2024-10-10 22:14:48, user Delphine Destoumieux-Garzon wrote:

Now published in PNAS: <br /> Oyanedel D, Lagorce A, Bruto M, Haffner P, Morot A, Labreuche Y, Dorant Y, de La Forest Divonne S, Delavat F, Inguimbert N, Montagnani C, Morga B, Toulza E, Chaparro C, Escoubas JM, Gueguen Y, Vidal-Dupiol J, de Lorgeril J, Petton B, Degremont L, Tourbiez D, Pimparé LL, Leroy M, Romatif O, Pouzadoux J, Mitta G, Le Roux F, Charrière GM, Travers MA, Destoumieux-Garzón D. Cooperation and cheating orchestrate Vibrio assemblages and polymicrobial synergy in oysters infected with OsHV-1 virus. Proc Natl Acad Sci U S A. 2023 Oct 3;120(40):e2305195120. doi: 10.1073/pnas.2305195120.

On 2024-10-10 07:25:37, user Flo Débarre wrote:

The points raised were dealt with in https://doi.org/10.1093/ve/vead077 . The paper was published after the preprint, but is cited and discussed in the peer-reviewed version of the work ( https://doi.org/10.1016/j.cell.2024.08.010 ).

As we do not believe that the comments made in this comment section have been in good faith, we will no longer reply.

On 2024-09-30 08:04:04, user Ema Nymton wrote:

Zach's reply does not really address the points in my previous comment. I note that it also does not address the points raised in Bloom 2024.

It only tangentially addresses 2 points in my post, both of which are apparently misconstrued.<br /> The first point that was addressed was about the p-values- however this response did not acurately portray the argument.<br /> My post never said it was misleading, but rather that the sampling bias towards the stalls would concentrate the highest p-values near the stalls, even more so than a simple relative-risk heatmap would, making the result of elevated p-values near the stalls expected whether or not the stalls were the origin of the outbreak.<br /> Compare: https://i.imgur.com/iDD6D93.jpeg to https://i.imgur.com/2YmYlZp.png

The second point "an argument that implicitly assumes all expectations for all coronaviruses are identical regardless of their hosts and modes of transmission,"<br /> No, it does not assume they are "identical", but rather that there are general similarities. I am asking for an explanation why the distribution is so different.<br /> Neither mode of transmission nor host explain this - nor does date of sampling.<br /> While not explicitly stated, there is a substantial focus on Raccoon dogs.<br /> You show the distribution of a virus with the same putative host (Raccoon dogs) and same transmission methods (respiratory). The distributions are strikingly different.<br /> Compare https://i.imgur.com/iDD6D93.jpeg to https://i.imgur.com/OFXpCIf.jpeg

This paper argued (in preprint), and it still argues )in final published form) that the 12 Jan RNA had more time to decay. Considering the market was closed at the same time, why did the SARS-CoV-2 reads decay to such lower values than other CoVs in samples from the same dates, and within the same samples, as shown by Bloom?<br /> See here: https://i.imgur.com/kIOTTYq.jpeg <br /> -even when stratifying by collection date, clear negative correlations still come up.

Overall, I find that a lot of speculation is offered, but it comes accross as excuses for the data not actually showing that the virus originated from the wildlife stalls.

I'll also raise additional critiques now:<br /> 1) The heat maps do not colocalize with the potential hosts:<br /> SARS-CoV-2 (RR p-values) and Raccoon dog reads shown here: https://i.imgur.com/AyAiag9.jpeg

In the paper, instead the focus is on the south-west portion of the market.<br /> As I see it, there are two possibilities:<br /> i) The analysis DOES have sufficient resolution: in which case the resulting heat map not co-localizing with the wildlife stalls would exonerate them<br /> ii) The analysis DOES NOT have sufficient resolution: in which case it must be noted that the eastern and northern parts of the market were very poorly sampled. In this case, finding more positive samples where more samples were taken does not allow any conclusions to be drawn.

2) The analysis of the SNPs of the Raccoon dogs clearly aligns with the the SNPs of locally caught wildlife (the C14859A + A15304G genome from Hubei, and the C14372T, C15102A, C15252A, T15306C genome from Hubei). Despite the evidence clearly pointing to locally caught raccoon dogs, speculation is offered (again) that maybe racoon dogs with these SNPs are found further south.<br /> Notably, Hubei (the local province) is not a plausible origin of the progenitor. Pekar seemingly agrees: https://www.biorxiv.org/content/10.1101/2023.07.12.548617v1

In essence, this paper shows:<br /> i) 3 lineage B sequences 1 low quality lineage A sequence have a (very very wide) tMRCA confidence interval that overlaps with a tMRCA calculated from many more A and B lineage genomes: hardly anything surprising or something that conclusions can be drawn from

ii) Within the heavily sampled area of the market, the areas of elevated relative risk don't actually overlap with the locations of the wildlife stalls<br /> -There is just one stall on the periphery of an area of elevated risk, the one closest to the entrance/mah jong rooms/bathrooms, which is where the higest elevated risk is.

iii) The raccoon dog genomes from that stall suggest that they are locally caught, and do not come from areas with potential reservoirs of SARS-CoV-2's precursors

On balance, the sum of this evidence seems to suggest that the HSM wildlife stalls in general, and raccoon dogs specifically, were not the origin of this virus.

Much speculation and many excuses are offered, but it must be reiterated that these excuses and speculation, not evidence.

The correlation analysis of Bloom can point right to the hosts of other viruses. The heat maps produced for this paper can point right to the hosts of other viruses. SARS-CoV-2 continuously fails to produce evidence and associations that other animal viruses at the market produced.

The way these findings are presented in the paper and to the general public through the media seems to be at odds with what the data actually shows.

On 2024-09-21 04:17:05, user Zach Hensel wrote:

I have a short response to the two comments on this preprint, which, of course, we took into consideration while revising the manuscript, which is now published following peer review here: https://doi.org/10.1016/j.cell.2024.08.010

One commenter, David Bahry, has taken to social media to call myself and co-authors "frauds" who are "trying to pretend" and made some vulgar comments that can't be repeated in this comments section because, he says, we "ignored" his comments. This is not true, so it's appropriate to respond here.

Both commenters note that low p-values for relative risk maps (Figure 2 and Figure 4) require sufficient sampling density to obtain a low p-value. Of course, this is correct. Both commenters argue that this is misleading. I disagree. No co-author, peer reviewer, or editor involved was misled. News articles on the paper thus far have almost all portrayed the data and analysis accurately. We do not argue in the paper that SARS-CoV-2 could not have been found in places that were not sampled; I'm unaware of anyone making this argument based on our paper; it's not misleading if no one is misled.

On top of this, the proportion of positive samples in each location is displayed along with the relative risk p-value. The n=1/1 sample that Bahry complains "shows little heat" is clearly indicated as a sampled location with high positivity in Fig 2A. And the underlying data is available in supplementary figures and tables. It's demonstratively not misleading.

The other arguments the two commenters make consist of (1) arguing against a different paper published years ago, (2) a demand for a citation of an irrelevant paper, (3) arguing for an alternative analysis method without demonstrating that it would have more statistical power, (4) an argument that implicitly assumes all expectations for all coronaviruses are identical regardless of their hosts and modes of transmission, and lastly (5) a typo in a citation. We addressed the last one -- showing that, in fact, we didn't ignore these comments.

On 2024-10-09 16:05:10, user Rosângela Maria Rodrigues Barb wrote:

Very interesting. I wonder if it works with urban pests (mosquitoes)

On 2024-10-05 16:00:52, user Walter S Leal wrote:

We are delighted to share this preprint on BioRxiv. It is the fruit of a couple of years of research collaboration with the citrus growers of the State of Sao Pauli. Fundecitrus has a unique model – they fund research (like most commodity groups do) and have PhD-level research scientists who engage in research activities. Please look at the number of replicates (as well as the raw data in SI) to further appreciate the work's depth. Walter Leal

On 2024-10-08 19:42:41, user Reena Sharma wrote:

Plant Microbe interaction is the epitome of resilience and symbiosis.

On 2024-10-08 19:41:38, user Reena Sharma wrote:

The key message of the study is that heavy metals like cadmium (Cd) and mercury (Hg) pose significant threats to plant health, but legumes, including Medicago truncatula, exhibit genetic variation in their ability to tolerate and accumulate these toxic metals. By conducting a transcriptomic analysis of plants with varying levels of tolerance to Cd and Hg, the study identified tissue-specific, genotype-specific, and metal-specific gene expression patterns.

Notably, plants inoculated with mercury-tolerant rhizobia strains carrying a mercury reductase (Mer) operon experienced less reduction in nodule number, plant biomass, and iron distribution under Hg stress. This suggests that Hg-tolerant rhizobia can mitigate Hg toxicity in plants, enhancing resilience in contaminated environments. These findings highlight the potential to optimize legume-rhizobia interactions for improving plant tolerance to heavy metals and reducing heavy metal transport to edible parts of the plant, which is critical for food safety.

On 2024-10-08 19:30:54, user Bogdan Olenyuk wrote:

This article is now published in Nature Nanotechnology. doi: 10.1038/s41565-024-01786-z

On 2024-10-08 18:44:34, user Paul wrote:

Please note that the preprint's result is subsumed by that in the published AMB paper. In particular, the AMB paper has modified Algorithm 1 so that it computes S in a right-to-left scan, rather than left-to-right. In this way, S does not need to be stored in memory, eliminating the quadratic space requirement. The space usage of the algorithm is now O(\ell \log \ell).

On 2024-10-08 09:43:09, user Bruno Cenni wrote:

Very nice and comprehensive dataset and overview across almost all BTKi. A note with regards to the data in Table 2 and Figure 4. For remibrutinib a BTK potency of 1.3 nM as “Kd or IC50” is listed. While the data is correctly referred to Angst et al 2020, this manuscript lists the IC50 for biochemical BTK enzyme inhibition. The same Angst et al 2020 publication also includes the Kd of remibrutinib for BTK (measured in the same assay all the others in the present manuscript) which was 0.63 nM. This is the value that should enter Table 2 and Figure 4.

On 2024-10-08 00:33:10, user Alexis Rohou wrote:

This comment is based on the version of the article published by Cell.

I have concerns about the validity of the cryoEM result presented in this paper.

The authors claim they obtained “a structure of the NINJ1 segment at 4.3 Å”. At this resolution, the following features should be clearly resolved in a cryoEM map of a protein containing alpha helices:<br /> • The pitch of alpha helices (~5.5 Å)<br /> • Large amino acid side residue side chains (e.g. Phe or Tyr, of which there are several in Ninj1)<br /> Neither of these features are demonstrated in the figures prepared by the authors.

Figure panels S2D and E are consistent with a resolution of ~ 8-10 Å, where alpha helices are resolved as tubular features. No regular indentation or ridging corresponding to the helical pitch is apparent in the figures, or upon visual inspection of the deposited map ( https://www.ebi.ac.uk/emdb/EMD-42301) . No features corresponding to Phe or Tyr side chains are visible. At the resolution claimed by the authors, features corresponding to the side chains of residues Phe100, Phe117, Phe127 and Phe135, which are all located in alpha-helical segments (not in loops), would be expected to be resolved in the map, but they are not.

In my view, based on this map inspection the authors should not have made this resolution claim.

To support their resolution claim, the authors present Fourier shell correlation (FSC) curves from the cryoSPARC software in Figure S2C.

While the FSC curve shown by the authors does cross the 0.143 threshold at ~4.3 Å, the FSC curve exhibits pathologies, which should have alerted the authors to the possibility that the 4.3 Å resolution estimate may be unreliable and that the map should be interpreted with caution. Most notably the curve starts dropping off at around 15 Å, indicating that the signal-to-noise ratio in the map is significantly deteriorated at resolution of ~15 Å and beyond.

After completion of 3D refinement, cryoSPARC also outputs a second FSC figure, which includes an additional curve (“Corrected”), which accounts for effects of masking on resolution estimation by the FSC. It is unfortunate that the authors didn’t include this in their manuscript.

The most likely explanation for this pathological FSC resolution estimate and its mismatch with the features resolved in the map is that the 3D refinement failed (due to high noise, or preferred orientation, or other pathologies in the dataset or in the refinement parameters), leading to overfitting to a local minimum in the scoring function.

Indeed, the validation report for the deposited map and PDB (EMDB: 42301; PDB: 8UIP; https://files.rcsb.org/pub/pdb/validation_reports/ui/8uip/8uip_full_validation.pdf ) contains evidence of overfitting during refinement. The orthogonal projections of the raw map (Section 6.1.2 of the validation report) show overfitting artefacts, as do the orthogonal standard deviation projections in false color of the raw map (Section 6.4.2).

Such overfitting causes artifically inflated FSC values, which may help explain why the FSC-based estimate of resolution was wrong in this case.

Given a map of this quality, it’s unclear how the authors built an atomic model of Ninj1. In that respect, the publication’s methods section is not detailed enough. Table S2 indicates that the authors started from a computational structure prediction from AlphaFold2. Given the lack of any features in the map to help place any residue side chains, I assume the location of key residues mentioned in the paper originated from this computational prediction rather than from the cryoEM result itself.

The lack of support for the modeled atomic coordinates from cryoEM is also made evident in the EMDB/PDB validation report by the unusually poor Q and atom inclusion scores for a map/model of the claimed resolution. A Q-score of 0.04 is unusually low and shows very little map-to-model fit.

These metrics, together with a visual inspection of the map and model as deposited, suggests that a significant fraction of the sequence was built outside the map and that the cryoEM result does not support the authors’ atomic model beyond the general shape and relative orientation of parts of the alpha-helical segments. The positioning of individual amino acid residues is not directly specified by the cryoEM result and may have come mostly from the AlphaFold model the authors used as a starting point.

Note: the above observations should be not be taken as having any bearing on other key cryoEM-based observations in the paper, such as the curvatures of Ninj1 assemblies, which are supported by two-dimensional class averages of cryoEM images and not affected by overfitting in the later 3D refinement process.

On 2024-10-07 14:40:02, user BindCraft Enjoyer wrote:

I like the ‘Design-Until’ architecture of the BindCraft pipeline, but one thing I couldn’t find in the paper is any quantification of BindCraft’s in silico design success rate. In the Introduction, you note that one drawback of RFDiffusion/MPNN-based pipelines is the need to screen thousands to tens of thousands of designs in silico before finding the 10-100 that pass the quality metrics and can be tested experimentally with good success rates. Does BindCraft also require screening of thousands to tens of thousands of designs, or is it more efficient in silico than an RFDiffusion pipeline? You mention in the paper that BindCraft outputs statistics from each design run, and that biasing away from alpha helical binders reduces the in silico design success rate; so it sounds like you have the statistics ready to hand, at least for the targets reported in the paper. I’d love to see these design success rates added to a table, either in the main paper or the SI.

Another thing I’d like to see is some quantification of the compute time and cost required to run the 4-step pipeline until 100 designs pass the in silico filters. I understand this cost scales with target/binder size and target difficulty, but I would imagine you have the data required to calculate these metrics at least for the design campaigns reported in the paper. I saw on Twitter that you’re working on a direct BindCraft / RFDiffusion pipeline comparison; I hope you’ll include the computational hardware and total CPU/GPU time for each side of that design campaign.

Great work!

On 2024-10-07 08:20:31, user Mike Pitcher wrote:

This article is now published with new title: "Double-negative B cells and DNASE1L3 colocalise with microbiota in gut-associated lymphoid tissue" in Nature Communications, 2024: https://www.nature.com/articles/s41467-024-48267-4

Michael Pitcher, co-author

On 2024-10-06 09:17:57, user Phil Jones wrote:

This article has now been published:<br /> https://www.nature.com/articles/s41588-024-01891-8

On 2024-10-06 09:16:37, user Phil Jones wrote:

This article has now been published as<br /> Self-sustaining long-term 3D epithelioid cultures reveal drivers of clonal expansion in esophageal epithelium,<br /> https://www.nature.com/articles/s41588-024-01875-8

On 2024-10-06 03:14:31, user David Grahame wrote:

I cannot find the references for the Supplementary material

On 2024-10-06 00:56:48, user Rishav Mitra wrote:

What is the endogenous TDP43 concentration in neuronal cells and how does that change with aging?

Are there neuron-specific gene regulatory mechanisms like alternative splicing, microRNAs, transcription factors, etc. that control TDP43 expression and consequently the intra-condensate localization inside SGs and liquid-to-solid transition?

Can microscopy as performed here distinguish between rapid coalescence of distinct TDP43 and G3BP1+ condensates and intracondensate localization?

Is the phenotype titrable with concentrationof arsenite, i.e., more number of intracondensate TDP43 foci at higher [arsenite]?

Does demixed TDP43 recruit/co-condense with other SG-resident proteins?

How can we correlate age-associated oxidative damage with the length of arsenite exposure?

Does solid-like demixed puncta of TDP-43 having higher local concentrations and slower internal dynamics compared to more liquid-like droplets potentially impact fixing and antibody staining? Is Correlative light and electron microscopy a better approach here than conventional immunofluorescence?

On 2024-10-05 08:38:14, user Martin GIURFA wrote:

Great work, congratulations! <br /> You may be interested in having a look at the following works, which relate to your findings:

° Pheromones modulate reward responsiveness and non-associative learning in honey bees. Baracchi D, Devaud JM, d'Ettorre P, Giurfa M. Sci Rep. 2017 Aug 29;7(1):9875. doi: 10.1038/s41598-017-10113-7

° Pheromone components affect motivation and induce persistent modulation of associative learning and memory in honey bees. Baracchi D, Cabirol A, Devaud JM, Haase A, d'Ettorre P, Giurfa M. Commun Biol. 2020 Aug 17;3(1):447. doi: 10.1038/s42003-020-01183-x.

Good luck with the next steps!

On 2024-10-04 16:24:29, user Gregory Way wrote:

We read this paper as part of a journal club, and have decided to compile a collective review and publicly share it with the authors. This was inspired by the Arcadia Science Preprint Review Pizza Party Initiative, and this represents our fourth preprint review.

Ji et al. present a transformer-based foundation model, called Prophet, which stands for Predictor of Phenotypes. The authors train Prophet on a variety of data modalities including gene expression, cell viability, chemical structures, and cell morphology using publicly-available sources. The authors should be commended for using such vast and disparate resources for such an innovative approach. The task of Prophet is to predict assay endpoints, such as cell viability, and to learn a useful embedding space which can be mined to identify novel, and potentially impactful, relationships. Most often, the phenotype prediction is in the context of some form of perturbation. The authors present a variety of benchmarks comparing Prophet to other methods, and they present both in vitro and in vivo applications to demonstrate potential use-cases. The applications range from looking up untested compounds that are similar to clinically-relevant compounds and predicting zebrafish cell type proportions after gene knockout. Overall, Prophet is methodologically interesting and the applications demonstrate that the method may help generate hypotheses at a low cost. However, we have several major and minor concerns mostly to do with clarity, performance, and software.

Major concerns:

1. Unclear and inconsistent terminology and definitions.<br /> a. It is unclear exactly what the authors mean by “phenotype”. It seems that sometimes the term is being used interchangeably with prediction/output but other times it is being used to describe observable physical properties. Additionally, the authors refer to gene expression as a phenotype, and, while technically true, it could be confusing given the authors are also using cell morphology as a phenotype. Furthermore, it is unclear if the authors are describing the collection of genes in the gene expression vector, for example, as the phenotype, or, if it is just a single gene. This confusion is also related to our confusion about model outputs (whether the output is a single element or a vector representation; see below). Given the word is in the article’s title, it seems particularly important.<br /> b. The terminology of “experiments” is also unclear. The authors claim to use 4.7 million experiments, but does this refer to plates, conditions, samples, something else? How did the authors calculate this count?<br /> c. At times, it is unclear what format the input data are and at what level of processing. What are the different possibilities of input data and how might a user decide which to use? Can a user input multiple kinds of data? Did the authors apply any sort of post-processing or quality control?<br /> d. The output of Prophet is ambiguous. Does Prophet predict a single value per input (or different inputs), or, does it predict a cell state vector? The authors describe outputting a one-hot encoding. Does this refer to the output phenotype? What is this structure? The authors write: “we train Prophet to predict cell viability, compound IC50, Cell Painting morphology features, mRNA transcript abundance, and cell type proportion.” Does this mean Prophet will output all of these predictions if the data you have is only morphology features? Does a user have control over these decisions? Furthermore, the authors write "Prophet’s transfer learning capability is not limited to phenotypes seen during the pre-training stage. We did not pre-train Prophet on any morphological measurement, but Prophet fine-tuned on JUMP outperformed both the Prophet-individual model trained only on JUMP and the baseline (Fig. 2b).” What is being predicted from the JUMP data? Morphology feature profiles? Images? Drug class or MOA? This needs to be expanded upon to make these claims. Please clarify the output structure and how a user will interact with the output. Figure 1C does not make this clear.<br /> e. Critical methodological details are discussed without sufficient detail. For example, the data split and validation strategies were ambiguous. How were training, test, and validation splits handled? What partitioning methods, if any, were used? The three-fold cross-validation procedure also lacked clarity. Were all datasets used in cross-validation? How did individual data-set models training differ and influence the full model fine-tuning? What is the specific pseudobulking procedure for RNA?
2. Unclear justification for Prophet architecture decisions<br /> a. The authors present table S3, which provides hyperparameters. How are these justified? For example, the choice of GeLU over alternatives like ReLU or SiLU. Why use an embedding dimension of 512? Were alternative configurations explored? How would modifying individual architecture decisions impact performance?<br /> b. 20.1 M parameters is a fairly small transformer, will this model need to grow as more perturbations or data types are added? In other words, how long will this model be foundational until the next best model is released? <br /> c. Encoders sequentially relate inputs with a positional embedding. Does this architecture use a positional embedding in the encoder? What does the position represent?
3. Concerns about model comparisons, baselines, and performance<br /> a. The authors compare Prophet to much simpler machine learning models (random forest, MLP, linear regression), individual Prophet models (trained using only one modality), and a mean baseline representing the average value of that intervention. The authors write: “This approach follows the same strategy as current foundation models (19, 36, 37), which are pre-trained on large amounts of data and then fine-tuned for specific datasets using the pre-trained model as a backbone.” Why not use these current foundation models as benchmarks for Prophet? The authors should also consider comparing different transformer architectures and non-transformer models (e.g., state-space models) as well.<br /> b. Prophet’s performance is low. The highest R2 value is 0.27 with a low of -0.03 and many predictions that perform the same as the mean baseline. Given the low, variable performance, it is difficult to trust Prophet’s output, or, at best, understand which outputs may have incorrect predictions. The authors claim an R2 improvement as low as 0.04 represents a 13x increase in number of hits, but it is unclear how the authors calculate this value. The authors also claim Prophet reduces “the number of experiments needed for viability screens by at least 60x” What statistics are calculating this estimate? <br /> c. The ML model comparisons compared to baseline are incredibly low. Results in Figure 2B suggest that the mean is a better predictor than nearly 100% of ML-based predictions (mean baseline is better in 41/45 comparisons). Our guess is that something might be going wrong in the model training or evaluation procedures.<br /> d. It is confusing if there is only one “final” Prophet model or if there are multiple “final” Prophet models since each must be fine-tuned to single datasets. Why not fine-tune on all datasets? Figure 2A suggests that for each time you perform inference in a new data category (e.g., cell morphology vs. gene expression), fine-tuning is required? If so, this strategy will lead to variable predictions and perhaps unexpected results and this is not a foundational model, but, at best, a foundational architecture.<br /> e. The authors performed a critical analysis, in which they restricted the amount of training data by 50%, 30%, 20%, 10%, and 5%. The authors state: “We found a clear trend: the more treatments and cell states seen by Prophet, the higher the confidence in the predictions (Fig. 2g).” This is an expected result, however, it is unclear how Figure 2G supports this statement. What is an “axis holdout”? Standard deviation of R2 for which predictions? What do the different points represent? How does this show confidence?
4. Concerns about limited discussion of technical artifacts<br /> a. The authors mention technical variation in the limitations section, but this should be elaborated upon. How might Prophet’s performance be impacted by technical artifacts?<br /> b. Earlier in the manuscript, the authors write: “We universally decompose each experiment into a unique combination of three fundamental elements—the cellular state, the treatments being performed, and the intended phenotypic readout.” Signals from technical artifacts are likely a fourth fundamental element, or, at the least, there should be an experiment to test the impact of technical artifacts.
5. Concerns about publicly-available source code on Github<br /> a. We provide a full GitHub review following our manuscript review below.

Minor concerns<br /> - Pg. 3 ln 5, the authors write: “To train it, we collected 9 perturbational datasets to create the largest compendium of publicly available screening datasets to date:...” The language suggests that these datasets were all collected by this group, but the datasets are all publicly available. Also, the Figure S1 reference probably should point to Figure S2. JUMP is listed three times in Figure S2 (one for compound, one for genetic treatment, and then one for both) Why is it listed for both when it is already split into the two perturbation types? Also, the right subplot in S2B (the complexity trade-off) is a bit misleading. JUMP has far higher complexity than PRISM, for instance, but this graph would suggest otherwise. Perhaps the missing piece not described here is readout complexity?<br /> - Figure 1B is a bit difficult to understand. For example, the bullet points for the “intervention” box don’t seem like interventions? It seems like these should be listed elsewhere or the heading should be changed. How is a “gene sequence embedding” an intervention data transformation?<br /> - The authors should tamp down claims. For example, the authors write “We did not pre-train Prophet on any morphological measurement, but Prophet fine-tuned on JUMP outperformed both the Prophet-individual model trained only on JUMP and the baseline”. The performance is only marginally improved (0.01). The authors retrained the zebrafish Prophet model, but it would be helpful to see performance for the original prophet model applied.<br /> - What is a pre-built plate? “To do so, we used a setup with existing pre-built 384-well plates, each with 352 unique drug perturbations applied to 9 cancer cell lines (Table S9). In total, there were 16 pre-built plates.”

GitHub comments and concerns<br /> 1. Documentation and Usability<br /> a. The README provided by the author is well-structured, offering clear instructions on installation, usage, and licensing, which provides a strong starting point for using Prophet. This level of clarity is especially valuable for researchers and users who are new to the tool. <br /> b. While the README provides a good overview, the documentation around model training is sparse. It would be beneficial to include an explanation in the README on how the model was trained and provide a small explanation of the embeddings captured. More detailed usage/inference examples would enhance comprehension. It’s also unclear how users can apply the method to their own datasets. This will offer quick and easy access for users to understand the functionality and purpose of Prophet.<br /> c. Commented-out code and TODOs are scattered throughout the scripts. Best practices suggest removing unused code to reduce clutter and confusion. Example: prophet/ http://model.py #L12.<br /> 2. Software Environment and Dependencies<br /> a. Users may face reproducibility challenges when trying to set up the software, particularly due to missing environment isolation. Creating an isolated Conda environment or improving instructions around environment setup would help ensure users avoid dependency conflicts.<br /> b. The project uses an older version of Pandas (1.5.x), despite newer versions being available with important fixes. Updating the Pandas version would improve compatibility and performance.<br /> c. A http://setup.py and requirements.txt are both provided but are not used together, creating confusion over proper environment management.<br /> d. No Python version range is specified in the http://setup.py , which led to issues with earlier Python versions (3.8, 3.9). Python 3.10 worked, but this should be clarified for future users.<br /> e. The provided notebook requires a specific version of NumPy that differs from what is stated in http://setup.py . Errors occur with newer versions. NumPy 1.24.4 was found to work, but this should be addressed in the dependencies.<br /> f. We were able to install it into a Linux machine. However, an error occurs when attempting to install the software on macOS. The error reports: “ERROR: No matching distribution found for scipy==1.14.0”<br /> g. A Jupyter notebook is included, but Jupyter is not listed as a dependency in the http://setup.py or requirements.txt, which prevents seamless execution within the provided environment.<br /> 3. Other comments<br /> a. The repository does not include any software tests or automated testing via GitHub Actions or similar tools. Incorporating automated testing would help validate the code’s functionality and improve its robustness.<br /> b. The code does not pass several linting checks (e.g., through dslinter), highlighting the need for improved code quality and adherence to data science best practices.<br /> c. The repository lacks key community health files like http://CONTRIBUTING.md and http://CODE_OF_CONDUCT.md , which are important for guiding open-source contributions and user interactions.<br /> d. Data provided in .xlsx format (e.g., via Figshare) can cause formatting errors and is less open-access friendly than formats like .csv. Switching to a more stable format would improve accessibility and avoid errors.<br /> e. In the tutorial notebook, the code blocks are not executed in a sequential order, which can lead to potential bugs. This lack of sequential execution means that changes made in earlier cells may not be reflected in subsequent cells, resulting in inconsistencies or errors in functionality. <br /> 4. Recommendations for Improvement<br /> a. Address installation issues: Fixing the bugs related to installation and setup should be a priority, as they could deter users from exploring Prophet further. Ensuring an isolated environment setup (e.g., using Conda) would help resolve these issues.<br /> b. Enforce version control for dependencies: Better organization of http://setup.py and requirements.txt, as well as enforcing version control for dependencies, would enhance reliability.<br /> c. Expand the README: Adding a table of contents, additional usage examples, and sections on contributing guidelines and testing procedures would make the README more comprehensive.<br /> d. Adopt best software practices: Implementing clear setup instructions, enforcing dependency version control, and organizing the code more effectively would increase usability and accessibility for a wider scientific audience.

This is a signed review:<br /> Gregory P. Way, PhD<br /> Erik Serrano<br /> Jenna Tomkinson<br /> Dave Bunten. MEd<br /> Michael J. Lippincott<br /> Cameron Mattson, MSc<br /> University of Colorado Anschutz Medical Campus, Department of Biomedical Informatics

On 2024-10-03 21:49:13, user Francesco Del Carratore wrote:

At the end of the methods section it is written 'To facilitate reproduction of these findings, all shareable data and code are available in a single structured file, with instructions and links for the non-shareable data, in S1 Data.'. This is great, but where can I find the S1 data as well as the code used for the analysis and figures (S1 code and S2 code)?

On 2024-10-03 21:10:43, user Gino wrote:

if I may ask, will this paper ever get out of the preprint stage?

On 2024-10-03 19:22:07, user Kaitlin Tagg wrote:

Hi bioRxiv!

This paper has now been published under a different title. Please see published version here: https://journals.asm.org/doi/full/10.1128/msystems.00365-24

Thank you,<br /> Kaitlin Tagg

On 2024-10-03 07:45:33, user Michiel wrote:

Interesting article on relevant topic. The recent article Identification, Elucidation and Deployment of a Cytoplasmic Male Sterility System for Hybrid Potato is a additional useful resource on this topic: https://www.mdpi.com/2079-7737/13/6/447

On 2024-10-02 20:15:19, user Prof. T. K. Wood wrote:

1. The authors fail to cite the relevant literature:

a. Line 59: the first report of ROS inducing persistence for any species is doi:10.1111/j.1751-7915.2011.00327.x (published in 2012, 12,000-fold increase) and should be cited. An erratum was issued to ref 20 by the Conlon group for failure previously to cite this reference.

b. The first high-throughput screen (10,000 compounds) for waking persister cells was doi:10.1111/1462-2920.14828 (published 2020), which reports persister cells wake by modifying ribosomes with pseudouridine.

c. The first report showing lower ATP increases persistence is doi:10.1128/AAC.02135-12 (published in 2013) and should be cited.

1. It is odd that the S. a. strains have 10% persistence (Fig. 1AB) when most strains have persistence far less than 1%.

On 2024-10-02 18:44:54, user Yasas Wijesekara wrote:

Congratulations on your discovery! This indicates that there is still so much biology waiting to be discovered. I'm interested in examining the genes on these Inocles myself. However, using the provided accession number, I couldn't locate the sequences. Would they be available soon?

On 2024-10-02 16:40:25, user Iva Tolic wrote:

This manuscript is related to our manuscript "Kinetochore-centrosome feedback linking CENP-E and Aurora kinases controls chromosome congression," doi 10.1101/2024.09.29.614573. This manuscript here expanded significantly during the revision from version 1 to version 2. As a result, we decided to divide it into two separate manuscripts: version 3 (doi: 10.1101/2023.10.19.563150v3) and a new manuscript (doi: 10.1101/2024.09.29.614573).

On 2024-10-02 16:36:02, user Iva Tolic wrote:

This manuscript is related to our manuscript "CENP-E initiates chromosome congression by opposing Aurora kinases to promote end-on attachments," doi 10.1101/2023.10.19.563150v3, which expanded significantly during the revision from version 1 to version 2. As a result, we decided to divide it into two separate manuscripts: version 3 (doi: 10.1101/2023.10.19.563150v3) and this one (doi: 10.1101/2024.09.29.614573).

On 2024-10-02 09:06:19, user Frédérique Reverchon wrote:

We discussed this paper in our Journal Club and the comments arising from our Discussion are the following:<br /> 1) Why focus on “rhizosphere soil” rather than “ bulk soil” to study the dynamics of fungal functional groups in fields of different age? The rationale behind that choice is not clearly explained, most of the references used in the Introduction and Discussion sections to explain the effect of time since planting on functional groups come from soil studies rather than rhizosphere studies. The term “field age” is also misleading as it suggests that bulk soil was sampled, and other terms such as “plant age” could maybe be considered.<br /> 2) There were no replicates per field age, as only one field per age was considered, although we understand that different plots were sampled per field. Other parameters than “field age” could cause the observed effects on fungal functional guilds, such as differences in soil nutrients between fields, for example. As seen in the PCoA, fields C and D (both classified as “old fields”) are different in terms of fungal community structure. Including replicate fields would have allowed to avoid possible confounding effects.<br /> 3) Important details are missing from the field description. Considering the Discussion on the possible effects of fertilization and fungicide applications, information should be added on the agrochemical management of these plots. Was it similar across fields? Across sampling years? Information regarding the soil nutrient status, and possible climatic variations between the two years of sampling, should also be included.<br /> 4) Methodological bias was also discussed. For example, most fungal taxa identified with the UNITE database would not be able to be assigned to a functional guild. How did you account for these taxa that could not be included into a functional category? Another possible bias that was discussed is the fact that relative abundance data may not represent what is actually occurring in the rhizosphere: absolute abundance data would also be needed and should be used for the implementation of microbial networks.<br /> 5) The discussion could also benefit from adding some consideration on the “functional group” categorization. How categorical is a “functional group”? Mortierella for example is listed as a saprotroph but it can also act as a plant pathogen. Mycorrhizal fungi included both EMF and AMF, yet strawberry only form symbiosis with AMF. As the focus of this study is on strawberry rhizosphere, why not discuss possible strawberry pathogens and symbionts?<br /> 6) An effect of plant genotype was not detected on relative abundance nor richness of fungal functional guilds. However, significant effects may be found in terms of taxa composition. Why not include such analysis in the study?<br /> Other minor comments are enlisted below.

Abstract<br /> Results seem to report differences in soil fungi rather than rhizosphere fungi. Why the focus on rhizosphere and not bulk soil? Plant genotype is not mentioned in the Results either, yet it was mentioned in the Background section as an important determinant of microbiomes.

Introduction<br /> L50: “The essential functions of soil microbiomes, such as pathogenesis, mutualism, and decomposition…”. Is pathogenesis a function? Functions of the soil microbiomes could rather include maintenance of soil structure, fertility, plant productivity… <br /> L66: Research gap: “Despite their significance, our knowledge of the dynamics of different fungal functional groups that co-exist in rhizosphere soil microbiomes is limited (Zanne et al. 2020; Martinović et al. 2021)”. The importance of filling this research gap is not clearly explained. <br /> L69: “Soil fungal functional groups can be dynamic in the plant rhizosphere. For example, fungal pathogens can accumulate over time, especially in monocultures or ecosystems with low plant species diversity (Cook 2006; Li et al. 2014; Peralta et al. 2018; Wang et al. 2023a)”. These references are for soil, not rhizosphere.<br /> Second paragraph of the Introduction: mostly on soil build-up of pathogens, and possible decrease of AMF over time. But this section seems to address the effect of long-term cropping, not plant age (rhizosphere). The rationale behind the selection of rhizosphere as the soil compartment to be studied is needed. You may want to check paper by Sun et al. (2022), published in Applied Microbiology and Biotechnology, on the dynamics of fungal functional guilds in the rhizosphere of wheat.

Methods<br /> Why establish and sample plots (20 x 20 cm) and not plants if rhizosphere soil was considered? Is there only 1 plant per plot (i.e. 3 plants sampled per genotype and field age)? It is unclear.<br /> Were the same plants sampled in 2021 and 2022? If not, a plant effect could be observed between both years.

Results<br /> L205: The increase in pathogens in the older fields is observed for one sampling year only (the pattern does not hold in 2022), yet it is described (and discussed) as a general finding.<br /> What happened in 2022 with AMF that decreased dramatically in all fields (both in relative abundance and in richness)? It is briefly discussed as possibly caused by the fertilization regime, which is not described.<br /> Differences in community composition could be assessed statistically to test for significant differences in taxon relative abundance between fields or sampling years. <br /> Regarding your Figures, stating the age of the field in the “x” axis rather than letters “A”, “B”, etc., could be more informative.

Discussion<br /> The discussion on the accumulation of pathogens over time is also based on references on soil microbiomes, not rhizosphere microbiomes. Were the same plants sampled in 2021 and 2022?<br /> L301: fungicide applications are mentioned to explain the “levelling off” of pathogens in older fields (C and D). Any data on application rates and application dates? This is important if fungal communities are studied. How specific were these agrochemicals? How were they applied?<br /> L311-312: Fusarium is a large genus with many species, not all are pathogens. Making inferences such as “the causal agent of Fusarium wilt in strawberry” is a bit far-stretched. Which Fusarium species are actual strawberry pathogens?<br /> The whole discussion regarding the potential accumulation of pathogens in the rhizosphere calls for symptom incidence data. It is something that was measured?<br /> L387: “the similarity in fungal functional groups across genotypes observed in this study may indicate functional redundancies and stable essential functions of microbiomes.” The microbiome was not studied and differences may be reflected in bacterial communities rather than fungal ones.

On 2024-10-01 20:16:33, user Beiyan Nan wrote:

This paper has been published in eLife https://elifesciences.org/articles/99273

On 2024-10-01 18:51:48, user Maria Valle wrote:

This review was done as part of the SfN Reviewer Mentor Program (Mentor: Joanne Conover, PhD; Mentee: Maria Luisa Valle, PhD)

Manuscript title: Human iPSC-derived pericyte-like cells carrying APP Swedish mutation overproduce beta-amyloid and induce cerebral amyloid angiopathy-like changes<br /> Journal: bioRxiv

Overview<br /> Wu et al. characterized human induced pluripotent stem cell (iPSC)-derived pericyte-like cells (iPLCs) to investigate the role of pericytes in Alzheimer’s disease (AD) and cerebral amyloid angiopathy (CAA). First, the authors showed that iPSCs could efficiently differentiate into pericyte-like cells, express pericyte specific markers, and promote angiogenesis, barrier integrity and contractility. They then investigated the differences between iPLCs derived from healthy individuals and those derived from AD patients carrying APPswe mutations. Compared to controls, APPswe iPLCs exhibited a distinct expression of pericyte markers, were able to secrete amyloid beta 1-40 and 1-42 within the media, had an altered transcriptome for key genes involved in cytoskeleton reorganization and metabolic regulation, were more sensitive to mediators of inflammation, and showed compromised angiogenesis, barrier integrity and hypercontractility.<br /> Overall, the manuscript uses a novel approach (iPLCs) to investigate an interesting and sometimes overlooked topic - the specific contribution of pericytes to AD pathology and vascular disfunction. Previous work conducted in in vitro BBB models showed that pericytes play a key role in amyloid clearance contributing to the removal of aggregated Aβ from brain capillaries (Ma et al, Mol Neurodegeneration 13, 57 (2018) and Blanchard et al, Nat Med. 2020 Jun;26(6):952-963). Here, the authors focus on the contribution of pericytes in amyloid secretion, emphasizing the novelty of their research. However, the high variability within datasets and the small number of replicates raises some concern.

Major comments <br /> • The statement “…overproduce beta-amyloid” in the manuscript title suggests that pericytes have a significant role in Aβ production. Although the authors showed that APPswe iPLCs could secrete 10 times more Aβ1-42 than the control cells, the Aβ1-42 levels are 100 times lower than neurons. Thus, the authors concluded that “contribution of pericytes to total brain amyloid load in AD is limited”. The title should be changed to indicate the main findings of the work and should be supported by the data presented. <br /> • APPswe iPLCs were derive from 3 donors versus iPLCs from 7 healthy controls. Importantly, among the donors, only one had AD, while the others had pre-symptomatic AD or no symptomatology (in this case the mutation was introduced using CRISPR-Cas9 as reported in the methods). The variability in AD cases plus the differences in symptomatology may skew the results and may contribute to the high variability shown in several graphs (Figure 2 A, B, C, F, J).

Figures<br /> • The authors should be consistent in the number of replicates used: different groups in the graphs show only 1 or 2 replicates, even for control cell lines, which makes the reader question the reproducibility and accuracy of their findings (see Figures 1B, 1I, 1K, 2H, 2J, 4E). <br /> • The authors should clarify the findings reported in Figures 2E and 2H: the figures are similar, but it is not clear if iPLCs in 2H derive from APPswe iPLCs (as reported in the figure legend) or control.<br /> • The authors should correct Figures 1A and 2I as sample labels are missing. The authors should also modify the arrows used in Figure 2I and 2D as it is not clear to what they are pointing. Scale bars should be added on both images since they show different magnification. <br /> • Figures should be arranged in a consistent manner e.g., same format and order should be used consistently.

Discussion <br /> • An interesting finding is that the HIF1a pathway is downregulated in APPswe iPLCs (Figure 3B). The authors should mention this finding in the discussion. This finding could also support the fact that APPswe cells have decreased VEGF levels and impaired angiogenesis and no change in BACE1 levels (as VEGF and BACE1 are HIF1a target genes). <br /> • For future experiments, the authors should discuss whether APPswe iPLCs exhibit differences in oxidative stress, ROS production and mitochondrial activity compared to controls.<br /> • For future experiments, the authors should use cell lines and human-derived cells as models as they may reveal differences from iPLCs.

Minor comments<br /> • The authors measured the changes in expression of several pericytes associated genes in Figure 1. However, it is not clear why the authors were not consistent with these specific genes for their further analysis. For example, in Figure 1B they measured PDGFRB, DES, LAMA2, DLC1, and PDE7B while in 1C they measured LAMA2, PDE7B, DES, omitting PDGFRB but adding genes ACTA2 and CD248. Then, all genes were analyzed in Figure 2A-B. Thus, the authors should provide change in expression data for all genes (PDGFRB, LAMA2, DLC1, CD248, PDE7B, ACTA2, DES) in 1B and 1C or provide reasoning for leaving some out. <br /> • Please correct the repeated sentences on page 5: “…which are known to express detectable levels of LRP121 (Figure 2 J). Furthermore, when iPLCs were subjected to pHrodo-conjugated zymosan-coated beads, no uptake of these pathogen-mimicking particles was observed (data not shown). Thus, it appears that the phagocytic activity of these iPLCs is low.”<br /> • Additional edits for word choice and sentence construction are also needed, e.g., pg 10, 2nd to last paragraph, 2nd to last sentence is awkward.

Decision for the editor: Major revisions<br /> The manuscript presents a novel idea that could advance the AD/CAA field but, at this stage, I have several major concerns regarding reproducibility and possible accuracy of the described findings. I would consider the manuscript for publication only if all major concerns are addressed by the authors.

On 2024-10-01 11:27:41, user Ana Matoso wrote:

This preprint has been published. You can see it in the following link: https://doi.org/10.1186/s10194-024-01854-8

On 2024-09-30 21:25:22, user Swagatam Barman wrote:

While the use of a massively parallel combination screen to identify an adjuvant for enhancing the efficacy of rifampicin is innovative, its practical implications are limited. Given that rifampicin is primarily restricted for treating tuberculosis, the likelihood of translating this adjuvant-rifampicin combination into clinical practice is quite low.

To improve the relevance of this research, it might be more beneficial to explore alternative drug combinations or focus on broader-spectrum agents that could be used in various clinical settings. This approach would enhance the potential impact of the findings and increase the likelihood of clinical applicability.

On 2024-09-30 15:40:53, user Christopher Dunn wrote:

I left a comment earlier, but it doesn't appear, so I am trying again.

This is an interesting paper, but I am not sure how wide an interest it will achieve. That aside, I need to re-read this and provide more detailed comments.

That said, I would note that the official state flower of Connecticut is not the marvel of Peru (Mirabilis jalapa). That species is the State's "Children's flower."

The official state flower of Connecticut is Kalmia latifolia (mountain-laurel). This is the species that should be used in the analysis.

On 2024-09-30 06:04:22, user honglab wrote:

We thank Dr. Imai for posting your results and hope this detailed protocol based on our paper can help all interested labs in achieving optical transparency in live mice with tartrazine: https://docs.google.com/document/d/1Da2WLIMZv0UOhV5Iesktp4mVkU5d_qb4wTGQKbPPeK8/

On 2024-09-28 08:57:40, user Stefania Kapsetaki wrote:

Now published in Evolution, Medicine, and Public Health doi: 10.1093/emph/eoae011

On 2024-09-27 22:21:40, user camilla wrote:

Final revised version is now available through eLife.

https://doi.org/10.7554/eLife.89543.2

On 2024-09-27 17:21:46, user Frank wrote:

After the CcFV-1 paper, which provided rather weak evidence for the virus being capsidated (no infectivity in addition to the rather weak blotting experimental designs), I find it difficult to agree with the wording of "strongly support the notion that polymycoviruses are encapsidated" being used by the authors again this time. Albeit the evidence is stronger here, there remain other plausible interpretations.

On 2024-09-27 12:56:53, user Morten Kjos wrote:

Published paper: https://journals.asm.org/doi/10.1128/mbio.01157-24

On 2024-09-27 11:24:18, user Ruth Berger wrote:

Important, high quality research. There is a potential alternative explanation for the pattern observed which should be checked: Viola arvensis is a pretty rare wildflower where I live, and in any peri-urban environment must be much, much rarer than garden cultivar violas that usually don't offer any pollinator food at all. Could it be that pollinators avoid them because of their experience with garden variety violas that taught them viola-like flowers are useless? From observation, I have seen various wild and cultivated viola species not getting any pollinator visits at all despite pollinators busily visiting other flowers in the immediate vicinity.

On 2024-09-27 08:42:12, user Hugo Amedei wrote:

Our articles has been published 2 month ago, ( https://www.frontiersin.org/journals/analytical-science/articles/10.3389/frans.2024.1425190/full )

On 2024-09-27 03:35:04, user fischmidtlab wrote:

Please find the final version of the manuscript here:<br /> https://www.nature.com/articles/s41467-024-52110-1

On 2024-09-26 21:44:42, user Lauren Porter wrote:

Most of the contents of this preprint have been included in this publication: https://www.nature.com/articles/s41467-024-51801-z

On 2024-09-26 20:35:22, user Trịnh Gia Huy wrote:

Hi Lacle, thank you for your comment. We fixed the figure already. The completed version and the code will be released later.

On 2024-09-26 18:45:20, user JC wrote:

Now published https://pubmed.ncbi.nlm.nih.gov/39266564/

On 2024-09-26 15:09:36, user pLM Enjoyer wrote:

An important application of pLMs is enhancing the efficiency of protein engineering by adding a classifier top model onto a foundation pLM (with or without fine-tuning), training on a small number (0-96) of experimental sequence/fitness datapoints, and then using this model to score and predict high-fitness sequence variants. This task also provides a good benchmark of pLM quality, since pLMs with ‘better’ embedded representations of sequences produce better variant scores/suggestions. See, for example, Zhou et al 2024 (Enhancing efficiency of protein language models with minimal wet-lab data through few-shot learning), and Jiang et al 2024 (Rapid protein evolution by few-shot learning with a protein language model). I think it would be really useful for you to benchmark AMPLIFY’s performance against ESM/SaProt/etc on few-shot and zero-shot variant fitness prediction using public deep-mutational scanning datasets as described in the papers above. If AMPLIFY really outperformed ESM2-15B on this task, that would be huge!

On 2024-09-26 14:13:00, user Dave Grainger wrote:

Hi, thanks for the great tool. I was looking through your figures for a journal club. I think there may be a slight error in panel c of the figure. The dotted outline does not highlight the same region in the b2c image as it does for 8um and MERFISH. Worth checking and updating for the final manuscript. BW, Dave

On 2024-09-26 13:55:29, user Paola Casanello wrote:

Very interesting data!<br /> We showed some years ago that adiponectin had vasodilator effects in the chorionic arteries (wire myography), and that the offspring of women obesity had a limited vasodilatory response to adiponectin.<br /> It would be interesting to discuss these results with your data.<br /> Muñoz-Muñoz et al., J Cell Physiol 2018. doi: 10.1002/jcp.26499.

On 2024-09-26 10:37:19, user Satoru Iwata wrote:

This preprint is published in Genetics: https://doi.org/10.1093/genetics/iyae054

On 2024-09-26 02:21:06, user Jeff Holst wrote:

Manuscript has been accepted for publication at EMBO Journal (24th September 2024). We will post a link to the open access revised version once it is available online.

On 2024-09-26 01:06:46, user Kate wrote:

This is the first author commenting here. Not sure where to discuss after the paper was published, but wanted to add some insight regarding the endogenously purified Pks13 protein used for cryoEM, crosslinking mass spec, and LC-MS (for identification of endogenous lipid). As it's published at NSMB, the paper doesn't address the fact that the endogenously purified Pks13 used for the above mentioned experiments, using the same purification protocol each time, showed variable SEC traces. After a while, we could not reproduce the peak1 and peak2 peaks in SEC (shown in Extended data fig1), but the two peaks either overlapped, or peak1 essentially disappeared. Sometimes, other interacting proteins were also pulled down along with Pks13, which confounded the SEC traces. CryoEM was done with peak1 with TAM inhibitor added, while XL-MS and LC-MS were done with mixed peak1&2 species. Protein used for XL-MS also had "contaminant" species which could actually be functionally-relevant binding partner. Since these protein preps behaved differently in SEC, we are not sure about the functional ramifications of these proteins purified at different times. If others are trying to reproduce our purification results, they may come across these variabilities (which is reasonable, given that this is an endogenous purification!).

On 2024-09-25 14:57:42, user Isabel Pires wrote:

This is now been published, see link below:<br /> https://pubs.rsc.org/en/content/articlelanding/2024/lc/d4lc00512k

On 2024-09-25 14:40:35, user David wrote:

What a great paper and story! It raises so many questions about the physiological relevance of such a mechanism specifically displayed by the gonadotrope cells! We were particularly interested to see that you identified Neurod1 and Neurod4 as being upregulated during postnatal differentiation of gonadotrope cells.

As we have recently shown in vitro (PMID: 37658038) that NEUROD1 and NEUROD4 regulate the mobility of immature gonadotrope cells by regulating the expression of NTRK3, we were wondering if you could identify this gene as well? It would be very interesting to know whether the mechanism that we have identified in vitro and in vivo as regulating gonadotrope positioning in the developing pituitary might be relevant to the process that you have just described.<br /> David and Charles

On 2024-09-25 06:52:59, user Qiang Kang wrote:

We have uploaded the latest pre-print version of manuscript on bioRxiv. And this manuscript has recently been accepted and published in Briefings in Bioinformatics ( https://doi.org/10.1093/bib/bbae450 ).

On 2024-09-24 16:44:20, user The Fehr Lab wrote:

These authors have done a fabulous job at creating new macrodomain inhibitors, which is extremely appreciated. However, having a major conclusion and an implication that Mac1 inhibitors are not antiviral based solely on negative data is misleading. We have published that a macrodomain inhibitor can inhibit virus replication (PMID: 38592023) and will have another story that will soon be available in BioRxiv that describes even more Mac1 targeting compounds that inhibit virus replication. There are some notable problems with the compound described here that could explain its lack of antiviral activity that should be taken into account. Again, I think the novel chemistry identified in this paper is exceptional, but more cautiousness should be taken before making broad claims that this is not a good drug target. Based on genetic data Mac1 and another recent paper on bioRxiv ( https://doi.org/10.1101/2024.08.08.606661 ), it appears that Mac1 is a suitable target for antiviral development, and we are continuing to work to see that dream come to fruition.

On 2024-09-22 21:30:29, user Christian Helker wrote:

Beautiful work!!! :)<br /> I would like to bring to your attention our publication (“Apelin signaling drives vascular endothelial cells toward a pro-angiogenic state”; https://elifesciences.org/articles/55589) , which explores the function of Apelin on the vasculature. I believe it could provide additional context or complementary insights to your work.

On 2024-09-22 20:05:49, user Fraser Lab wrote:

Based on: https://www.biorxiv.org/content/10.1101/2024.07.24.604935v1.full

The manuscript from Lehner and colleagues presents a wealth of mutagenesis information on amyloid aggregation. The central premise of the paper is to use a yeast selection based around the oligmerizaton/aggregation of Sup35 fused to a peptide (in this case abeta) as a proxy for amyloid forming potential. This is cool information on its own and the experimental analysis and computational framework for linking to energies is top notch. The point of using double mutants to enhance the dynamic range is very well explained and will solidify their approach to impactfully link DMS experiments to thermodynamic concepts.

The major framing of the paper revolves around an analytical protein folding/engineering concept of phi-values that highlights energetic differences in the importance of interactions for forming the transition state vs. the ground state. For the textual interpretation of the results, one must buy into the energetic effects of the Sup35 system as a readout of the transition state (and secondarily for FoldX calculations on various PDBs of abeta polymorphs to be a readout of the ground state). The major issue is that an alternative (and perhaps simpler) explanation is that mutations in APR2 are more disruptive to the Sup35 oligomerization process in the screen and that this reflects amyloid/oligomerization propensities and not strictly TS of initial nucleation. The data from previous studies that is used to draw correlations to justify their interpretation around the TS is buried in extended data figures and is a bit all over the place, especially the deconvolution of primary vs. secondary nucleation. The existence of multiple polymorphs in human cells (and populations), which may or may not have related transition states - and the exact conformational requirements of the Sup35 activation mechanism - further complicate this interpretation.

In summary - this contains amazing data, but I do not see the language of the interpretations lining up with the strength of the _specificity_ of the claims about the transition state. A fuller discussion of the limitations of the prior low throughput assays that are referenced in extended data 1 and 2 and detailed kinetic characterization of some of the more surprising mutants in a biochemically defined system would greatly improve the match between the data and the claims. These issues should not stop others from building on this beautiful work - but in doing so, other investigators should note that there remains ambiguity as to whether the effects are truly on the TS or on the ground state.

Avi Samelson and James Fraser

On 2024-09-22 19:30:05, user Subha Kalyaanamoorthy wrote:

The paper has been published in Nature Machine Intelligence. Here is the citation: Reeves, S., Kalyaanamoorthy, S. Zero-shot transfer of protein sequence likelihood models to thermostability prediction. Nat Mach Intell 6, 1063–1076 (2024). https://doi.org/10.1038/s42256-024-00887-7

On 2024-09-21 18:49:07, user Flo Débarre wrote:

Following up on my previous comments about the pangolin datasets featured in Figure 2:

It had been suggested that NCBI could have tampered with the dates and times shown on their systems. To confirm that a request had been made on June 16, 2021 to make the data public again, I FOIA'd NIH with more search terms than had been done in previous requests. I finally received an email with the user's request, and I can therefore confirm that the pangolin CoV data being public again is unrelated to Jesse Bloom's preprint having been sent to bioRxiv or to NIH on June 18.

On 2024-09-20 11:47:19, user Tomas Strucko wrote:

This preprint was published in the Metabolic Engineering journal with a title "Genome-wide host-pathway interactions affecting cis-cis-muconic acid production in yeast"<br /> https://doi.org/10.1016/j.ymben.2024.02.015

On 2024-09-19 20:21:02, user Danve Castroverde wrote:

Final peer-reviewed version has been published: https://doi.org/10.1111/pce.15098

On 2024-09-19 14:33:26, user Tomas wrote:

This preprint is now published in Nucleic Acid Research with a Title "CRI-SPA: a high-throughput method for systematic genetic editing of yeast libraries" https://doi.org/10.1093/nar/gkad656

On 2024-09-19 14:29:20, user Tomas Strucko wrote:

This preprint is now published in FEMS Yeast Research https://doi.org/10.1093/femsyr/foae026

On 2024-09-19 14:10:01, user Farhan Feroze wrote:

Excellent work!<br /> I am curious about the reasons why pulse code EO-151 was preferred over EH-115?<br /> Also, were whole plasmids used as a HDR templates for electroporation? (Since we usually deliver the HDRT as linear dsDNA or ssODN with exposed homology ends)

On 2024-09-18 21:12:56, user Jason Hoskins wrote:

I am a fan of this approach of pairing multidimensionality reduction of the expression data with biologically meaningful gene sets to generate representative scores of processes or pathways that may be used as quantitative traits in QTL analyses. Approaches like this applied to GWAS variants enable the implication of processes or pathways potentially mediating the genetic effects on complex phenotypes even in the absence of co-localizing cis-eQTL signals, which is unfortunately typical of GWAS signals.

This pre-print does not include a Discussion section, so it does not yet sufficiently place this work in the context of the relevant published literature. When this is done, I would recommend consideration of our work on expression regulator activity QTLs (aQTLs) that was published in PLOS Computational Biology a few years ago ( Hoskins et al., 2021 ). Our two approaches share a lot conceptually in that we both assume that shared variability among genes often reflects a common underlying regulatory mechanism and that multidimensional reduction of the expression information among such genes can provide a useful metric for summarizing the activity of the gene set. The approach in this pre-print is generalizable for use with any type of gene set deemed potentially relevant to the GWAS trait of interest, while our approach in Hoskins et al. (2021) is more tailored to gene sets representing the target genes of expression regulators based on tissue or cell type-specific co-expression networks where the activity scores are inferred using the VIPER algorithm developed by Andrea Califano’s group (Alvarez et al., 2016). There are also some analyses, results, and speculations in our paper that might suggest potential additional considerations and investigations for this gsQTL study.

On 2024-09-18 14:19:15, user Musaeum Scythia wrote:

I was curious on which basis the authors wrote the following section:

"One particularly intriguing finding is the identification of the Y-chromosome haplogroup Q1b1a3a1 in an individual from Santarém_Rua_dos_Barcos_13th-c (PT_23227). This haplogroup is rare in Iberia but more commonly associated with Ashkenazi Jewish populations and Central Asia."

Were there any sources used for this statement?

As far as I know, Jewish clades under Q are typically not Q1b1a3a1/Q-L332 (clades under Q-L245 are more typical).

Q-L332 is a Y-chromosome haplogroup found in several bronze age Siberian populations, and is a fairly prominent lineage across Scythian populations during the iron age and antiquity (hence my interest in the remark above).

Given the history of the Iberian peninsula, would it not be more plausible to attribute such lineages to the Alans? Sarmatians carried Y-chromosome haplogroup Q-L332, and there seem to be a decent amount of Spaniards and Portuguese which carry Sarmatian-related lineages to this day. Perhaps the R1a-Z94 lineage could also be there through Sarmatians but it would depend on the subclade, as R-Z94 is over 4000 years old and was carried by various peoples - mostly of Indo-Iranian origin however.

If there were any sources used regarding the Jewish origin of Q-L332 in the Iberian peninsula or perhaps archaeological signs which affirm the Jewish origin of sample PT_23227, I would be interested to have a look.

Thanks in advance!

On 2024-09-17 10:34:10, user balli wrote:

The authors state in the intro... "only one clinical study has been published...."<br /> Please look at Jebsen et al 2019 J Med Case Rep / Spicer et al 2021Clin Cancer res.<br /> Also, using short peptides restric their use for intratumoral administration, please provide evidence.<br /> What about immunogenicity and potential ADA responses by longer peptides, please adress.<br /> Given that BOP peptides are strong inducers of ICD, why was immunodefect mouse model used ?

On 2024-09-17 08:21:43, user EDG wrote:

This article is now published in Nature Communications : https://doi.org/10.1038/s41467-024-51638-6

On 2024-09-17 08:15:02, user Adriana Ludwig wrote:

Dear authors. Please check some issues identified with TransposonUltimate tool which I believe can have an impact on your findings: https://github.com/DerKevinRiehl/TransposonUltimate/issues/16

On 2024-09-16 12:23:28, user Jacques Pécréaux wrote:

This manuscript is now published by PLoS Comp Biol ( https://doi.org/10.1371/journal.pcbi.1012330 ). Please refer to this revised and augmented version.

On 2024-09-15 19:00:15, user Dr. M. K.Tiwari wrote:

This mycobacterium may be similar to M tuberculosis, but it doesn't seem to be very unnatural as M.tuberculosis like organisms are reported from number of aquatic animals, engulfment of mycobacteria by amoeboid cells is well known and these mycobacteria are not digested in food vacuole of amoeba but some time there multiplication in vacuole is also reported.<br /> Sponge may have engulfed the mycobacteria from ablutions of an infected patient and entered in sponge along with incurrent water in spongocoel these mycobacteria were ingested by choanocyte where probably these bacteria were not digested and kept on multiplying. So it may be a possibility that this is not the mycobacteria originally from sponge but entered in sponge along with debris or ablutions of infected patient.

On 2024-09-15 07:54:12, user PanosMoschou wrote:

We will soon test the approach using various super-resolution approaches down to 20 nm. Will keep you updated.

On 2024-09-14 15:54:22, user Juri Rappsilber wrote:

Now published: <br /> https://www.sciencedirect.com/science/article/pii/S0022283624002511?via%3Dihub <br /> https://pubmed.ncbi.nlm.nih.gov/39237202/

On 2024-09-13 21:17:20, user Bennett Fox wrote:

out in PNAS: https://www.pnas.org/doi/full/10.1073/pnas.2221150120

On 2024-09-13 21:13:30, user Bennett Fox wrote:

now published here: https://elifesciences.org/articles/58041

On 2024-09-13 15:32:48, user Ibrahim, Tarhan E wrote:

Chung et al. (2024) identified a physical interaction between ERC1 and ATG8e, leading them to explore potential ATG8-interaction motifs (AIMs) in ERC1. Using the iLIR database (Jacomin et al., 2016) for AIM prediction, they found the results irrelevant to the ERC1-ATG8e interaction, indicating a false prediction. Through truncated ERC1 variants, they identified a non-canonical AIM undetectable by current prediction tools, which focus on the canonical [W/F/Y]-[X]-[X]-[L/I/V] sequence. They validated this motif with AlphaFold2-multimer (AFM), a method we previously demonstrated (Ibrahim et al., 2023) to accurately predict non-canonical AIMs, as shown with ATG3. Our findings were later confirmed in humans by Farnung et al. (2023) via X-ray crystallography. Despite their similar approach, Chung et al. (2024) did not acknowledge our prior work

On 2024-09-13 14:53:14, user Thibaud Decaens wrote:

The manuscript has now been published in European Journal of Soil Biology:<br /> Gabriac Q., Ganault P., Barois I., Angeles G., Cortet J., Hedde M., Marchán D.F., Pimentel Reyes J.C., Stokes A., Decaëns T. (2023) Environmental drivers of earthworm communities along an altitudinal gradient in the French Alpes. European Journal of Soil Biology, 116, 103477. https://doi.org/10.1016/j.ejsobi.2023.103477

On 2024-09-13 06:17:44, user Massimo Turina wrote:

Interesting, but make sure you place it taxonomically in the new family "Konkoviridae".... therefore not really correct to call it a new "Phenuivirus"......but anyway good job.

On 2024-09-12 15:41:59, user Hyunjin Kwun wrote:

This preprint is published:<br /> Pham AM, Ortiz LE, Lukacher AE, Kwun HJ. Merkel Cell Polyomavirus Large T Antigen Induces Cellular Senescence for Host Growth Arrest and Viral Genome Persistence through Its Unique Domain. Cells. 2023 12(3), 380; https://doi.org/10.3390/cells12030380

On 2024-09-12 09:32:02, user Fabienne Jabot-Hanin wrote:

Could you please share your supplementary table 2 with the 385 index variants which had the same direction of effect on serum creatinine and cystatin C and had significant effects on both biomarkers ?

On 2024-09-12 07:42:51, user maud de dieuleveult wrote:

Published here https://www.tandfonline.com/doi/full/10.1080/15592294.2021.1917152

On 2024-09-12 07:11:20, user Keshava Datta wrote:

Great study - Extremely important to improve genome annotation as we know it... In one of the first drafts of the human proteome (Nature, 2014), ~16 million spectra that did not match to known proteome were subjected to proteogenomic analysis and ~200 regions with protein coding potential were found. It would have been great if the authors mentioned this and maybe compared these results? As we all know, evidence from multiple groups increases confidence in a finding!!

(Full disclosure - I was a co-author on the 2014 paper)..

On 2024-09-11 02:22:16, user Shinya Fushinobu wrote:

This paper has been published in the Journal of Applied Glycoscience.<br /> https://doi.org/10.5458/jag.jag.JAG-2024_0005