On 2015 May 29, Michal Kicinski commented:

I thank Dr. Hilda Bastian for her interest in our recent study (Kicinski M, 2015). I strongly believe that post-publication comments very often raise important issues and help the readers to better understand the merits of a study and its limitations. However, I was disappointed to see that the comments of dr. Hilda Bastian do not correspond with the content of our study. For this reason, I feel obliged to clarify a number of issues.

The study of Ioannidis JP, 2007 points out one of the limitations of a large part of publication bias methods based on the asymmetry of the funnel plot, namely that they do not take between-study heterogeneity into account. This is indeed an important limitation of these methods, as also discussed by other researchers (Song F, 2010). However, please note that we did not rely on the asymmetry of the funnel plot in our analysis. Additionally, please note that our model is just an extension of the standard random effects meta-analysis model, which is a valid approach when between-study variability is present. In fact, the study of Ioannidis JP, 2007 is one of the contributions that motivates our approach to model publication bias since our model takes heterogeneity into account.

Dr. Hilda Bastian correctly points out that our study is not the first study on publication bias. There are many valuable studies on this topic and we discussed those most relevant to our research questions in our article. The contribution of our study is that we analyzed a very large number of meta-analyses using a model with strong theoretical foundations. Our study is the largest study on publication bias in meta-analyses to date. Please note that previous studies, e.g., Ioannidis JP, 2007, which Dr. Hilda Bastian mentioned, considered small study effects, a phenomenon that may have many different causes, including publication bias (Song F, 2010, Sterne JA, 2011). Another merit of our study is that we estimated the association between the size of publication bias and the publication year of the studies included in the meta-analyses.

I completely agree that the best solution to the problem of publication bias is the complete reporting of study results. In fact, our findings showing that publication bias is smaller in the meta-analyses of more recent studies support the effectiveness of the measures used to reduce publication bias in clinical trials. I strongly advocate the introduction of new policies aimed to completely eliminate reporting biases from clinical trials and, as written in our article, the implementation of measures to reduce publication bias in research domains other than clinical trials, such as observational studies and preclinical research.

Although we did not investigate the use of publication bias methods in the meta-analyses from the Cochrane Library, it is clear from previous research that the potential presence of publication bias is often ignored by researchers performing meta-analyses and that the methods accounting for publication bias based on the statistical significance are hardly ever used (Song F, 2010, Onishi A, 2014). When publication bias is present in a meta-analysis, ignoring the problem leads to biased estimates of the effect size (Normand SL, 1999). Therefore, similar to others (Sterne JA, 2011), we argue that researchers should investigate the presence of publication bias and perform sensitivity analyses taking publication bias into account. One difficulty with the use of publication bias methods is that they require researchers to make certain assumptions about the nature of publication bias. For example, the trim and fill method defines publication bias as suppression of a certain number of most extreme negative studies (Duval S, 2000). The use of the Egger’s test (Egger M, 1997) as a publication bias detection tool requires researchers to make the assumption that publication bias leads to a negative association between effect size and precision. The performance of a certain publication bias method depends on whether or not the method’s assumptions are met. For example, it has been demonstrated that publication bias detection tests based on the funnel are characterized by a very low power when publication bias based on the statistical significance is present and the mean effect size equals zero (Kicinski M, 2014). Publication bias based on the statistical significance is the best-documented form of publication bias (Song F, 2009, Dwan K, 2013), The results of our study add to this body of evidence. Therefore, we argue that publication bias tools designed to handle publication bias based on the statistical significance should be used by researchers.

In the tweet with the link to her comment on PubMed, Dr. Hilda Bastian wrote on the 25th of May: ‘27% of cochranecollab reviews over-estimate effects cos of publication bias? Hmm.’ Please note that our study did not investigate the proportion of meta-analyses that overestimate effects. In fact, the objectives of our study were completely different. We estimated the ratio of the probability of including statistically significant outcomes favoring treatment to the probability of including other outcomes in the meta-analyses of efficacy and the ratio of the probability of including results showing no evidence of adverse effects to the probability of including results demonstrating the presence of adverse effects in the meta-analyses of safety.

On 2015 May 25, Hilda Bastian commented:

This is an interesting study. But it's a rather enthusiastic self-assessment of a method not validated by other researchers, and some perspective is useful in thinking about the conclusions.

Kicinski M, 2015 is neither the first, nor the largest study, of publication bias (PB) in meta-analyses, and the presence of publication bias in them is well-known. These authors used a scraper they have made available on Github to extract meta-analyses from Cochrane reviews. They looked at reviews with placebo or "no treatment" control groups and 10 or more included studies. Whether or not these results are applicable to interventions with active or usual care control groups is unknown.

For perspective here: Ioannidis JP, 2007 considered PB in 1,669 Cochrane reviews, ultimately analyzing 6,873 meta-analyses. A half of the meta-analyses had no statistically significant results in them, so the problem identified here could not have applied to them. Ioannidis JP, 2007 concluded that only 5% of the full set of Cochrane reviews would qualify for the use of asymmetry tests, and only 12% of those with a larger number of events and participants. They found very little concordance between different asymmetry tests - only around 3-4%. A more important problem according to Ioannidis JP, 2007 was the misapplication and misinterpretation of statistical tests, not under use. False-positives are a problem with tests for PB when there is clinical heterogeneity. Ioannidis JP, 2007 conclude that the only viable solution to the problem of PB is full reporting of results.

Kicinski M, 2015 conclude that statistical tools for PB are under-utilized, but the extent to which PB is assessed was not part of their study. Although PB itself may be decreasing over time, assessment of PB is increasing, even if the methods for exploring it are still problematic:

• Palma S, 2005 found that PB was assessed in 11% of trials between 1990 and 2002, increasing from 3% in 1998 to 19% in 2002 (less frequently in Cochrane reviews than others).
• Moher D, 2007 found that about 23% of systematic reviews in 2004 assessed PB (32% in Cochrane reviews, 18% in others).
• Riley RD, 2011 found that only 9% of reviews from one Cochrane group assessed PB.
• van Enst WA, 2014 found that most systematic reviews of diagnostic test accuracy in 2011/2012 mentioned the issue, with 41% measuring PB.

In assessing only the meta-analyses themselves, and not the reviews that included them, it's not possible to know, as the authors point out, to what extent other studies were included, but without data that could be pooled. An issue not raised by Kicinski M, 2015 are trials reported only in conference abstracts, and thus with minimal data. Cochrane reviews often include studies reported in conference abstracts only, and those are apparently more likely to have non-statistically significant results (Scherer RW, 2007) - as well as relatively little data for the multiple meta-analyses in a review.

It's important to consider the review, and not just the effect summaries within meta-analyses, because the conclusions of the systematic review should reflect the body of the evidence, not only the meta-analyses. Over-favorable results in a meta-analysis shouldn't be equated with over-favorable conclusions about effectiveness in a review (although unfortunately it often will). We shouldn't jump to conclusions about effect sizes from meta-analyses alone. They can be skewed by clinical heterogeneity and small study size as well as (or instead of) publication bias, and the devil may be more in the interpretation than the calculations.

Disclosure: I work on projects related to systematic reviews at the NCBI (National Center for Biotechnology Information, U.S. National Library of Medicine).

On 2015 May 26, Yanju Ma commented:

I am very interested in your study. In this study, you found no correlation between the numbers of CTC-clusters and that of single CTCs (Supplementary Fig. 9). But we could see that three patients with melanoma have higher CTC clusters. So is there any correlation between higher CTC clusters with poor prognosis? I will be glad to get your reply. Thank you.

On 2015 Jun 14, Preben Berthelsen commented:

In this editorial, Prof Matthay contents that the paper by Blodgett: The continuous inhalation of oxygen in cases of pneumonia otherwise fatal, and in other diseases. (Boston Med Surg J 1890;123:481-) is the first report on oxygen therapy. Blodgett’s case story is indeed interesting and heart-breaking but it is not a first – and Blodgett certainly does not claim priority in his paper.

Fifty years earlier, in “On Asphyxia, and on the Resuscitation of Stillborn Children”(a paper read at the Westminster Medical Society, London, October 16, 1841 and printed in Boston Med Surg J 1842;25:409-) John Snow states on page 414 “Oxygen gas is sometimes mixed with air to be thrown into the lungs of asphyxiated persons” And, “if it is deemed advisable, oxygen gas can be generated in great purity, in a few minutes, from chlorate of potash by means of a spirit-lamp and a small retort” Furthermore, “No harm can arise from thus using oxygen.”

I do not know if John Snow – the first academic anaesthetist in the world – was also the first to advocate oxygen treatment.

P.G.Berthelsen, MD. Charlottenlund, Denmark

On 2015 May 22, thomas samaras commented:

Researchers should be aware that two earlier reviews were published showing that shorter people are more likely to have lower coronary heart disease (CHD) (1,2). This apparent conflict with the Nuesch, et al. paper is no doubt due to various confounders and failure to recognize the biological mechanisms that underlie the inherently lower risk of CHD for shorter, lighter people. In addition, ecological studies showing short people often have little or no CHD are often ignored because of the belief that ethnic, racial, climatic and geographical factors make the results unreliable. However, Shaper (1) stated that these factors are completely irrelevant to the incidence of CHD unless they are related to social class and economic status.

In my opinion, the facts covered in the two papers cited below (1,2) provide overwhelming evidence that shorter people have the potential for very low CHD. (Unfortunately, Dr. Elrick and Professor Storms have passed away and cannot co-write this response.) A summary of some of our findings follows.

Before the early 1900s, Europeans and Americans were shorter but CHD was rare. As we got taller and heavier, CHD increased due to the Western diet.

Based on 1 million WWI military recruits in the US, Davenport and Love found that taller men had a much higher incidence of heart problems.

In the developed world, the Japanese have ranked within the top three countries in terms of low CHD for a number of decades. Japanese males average a little over 5’7”.

A number of populations that had very low or no CHD during the 20th C now have much higher rates along with their increased height. For example, S. Korea has seen a large jump in heart disease in recent years along with a substantial increase in height.

Western Sicilian centenarians are short and free of CHD risk factors.

Compared to men, shorter women have lower lifetime death rates from CHD.

A US study found that in a low-income cohort, taller people have almost a 40% higher risk of heart attacks. US mortality statistics based on ages ranging from youth to over 85 years of age show that shorter Asians have an almost 80% lower risk of CHD compared to taller Blacks and Whites. Latinos and Native Americans are taller than Asians but shorter than Whites and Blacks and fall between the shortest and tallest ethnic groups in mortality. (Findings based on a 15-year period and about 8 million deaths.)

A study of 350,000 dogs found that smaller dogs have much lower risk of heart failure compared to bigger dogs. For example, the Great Dane has 70 times the risk of heart failure compared to the miniature Dachshund. And the standard Dachshund has 3 times the rate of heart failure as the smaller miniature Dachshund.

I have also identified over 15 biological factors that would explain the advantages of smaller body size and low CHD. These include longer telomeres, smaller left ventricular mass, lower blood pressure, lower HDL, lower homocysteine, higher sex hormone binding globulin, lower C-reactive protein, higher adiponectin and FOXO3, and lower pulse wave velocity. These factors are based on the assumption that we are comparing shorter and taller people who have the same body type or proportions. (Otherwise, comparing a tall, thin person to a short stocky one would change these parameters.)

Many studies have also found that shorter people live longer. These include studies from the Hawaii, Ohio, San Diego, Spain, Sardinia, Okinawa, Bama (China), and Cuba (3). Many animal studies and experimental research support these human findings (4).

Recently, a 40-year longitudinal study based on over 8000 elderly Hawaiian Japanese males found that shorter men live longer (5). It is unlikely that they had more heart problems than the taller men who died younger.

Unfortunately, the findings discussed above have not received much attention because of the widespread belief that taller people are healthier and longer lived. Over 45 papers, books and book chapters on the ramifications of increasing body size are listed in www.humanbodysize.com

References

1 Samaras TT, Elrick H, Storms LH. Is short height really a risk factor for coronary heart disease and stroke mortality? A review. Med Sci Monit 2004; 10(4): RA63-76.

2 Samaras TT. Shorter height is related to lower cardiovascular disease risk—a narrative review. Indian Heart Journal 2013; 65: 66-71.

3 Samaras TT. Evidence from eight different types of studies showing that smaller body size is related to greater longevity. Journal of Scientific Research & Reports 2014; 3 (16): 2150-2160.

4 Bartke A. Healthy aging: Is smaller better? –A mini review. Gerontology 2012; 58: 337-43

5 He Q, Morris BJ, Grove JS, Petrovitch H, Ross, Masaki KH, et al. Shorter men live longer: association of height with longevity and FOXO3 genotype in American men of Japanese ancestry. PLOS One 2014; 9 (5) 1-8.

On 2015 Jun 22, Darko Lavrencic commented:

Your conclusions indicate that spinal extrathecal CSF collections are not the cause but direct consequence of intracranial hypotension/CSF hypovolemia, regardless of CSF leakage. For the most probable cause of intracranial hypotension/CSF hypovolemia see http://www.med-lavrencic.si/research/correspondence/

On 2015 May 29, Kim D Pruitt commented:

lncRNA-NR024118 is represented by NCBI Gene ID: 25602 and RefSeq accession NR_024118.1

On date unavailable, commented:

None

On 2015 Sep 23, Guangchuang Yu commented:

• 1. The so called Linux-like should be Unix-like.
• 2. you should also present runseq2pathway runnable example.
• 3. it seems you are not familiar with R

tmp <- sessionInfo() mySession <- ifelse(length(grep("Windows",tmp))==0, "L","W")

I recommend you use Sys.info() to extract platform information.

if (mySession=="W") sink(paste(tempdir(),"\",name,sep="")) else sink(file.path(tempdir(),name,fsep = .Platform$file.sep))

actually '/' works fine with Windows and Unix-like system in R. Why not just use:

name = tempfile()

and write your python script to that file?

On 2015 Sep 20, Xinan Holly Yang commented:

Thank you for pointing out. We have fixed the bug in the seq2pathway version >=1.1.6 as described below:

1) The new seq2pathway package runs on both Windows and Linux-like systems.

2) We activated a demon code:

data(ChipseqPeakdemo)

runseq2gene(inputfile=ChipseqPeakdemo)

3) We replaced the absolute path with Sys.which("python").

The significance of the package for the end-users could be:

1) It provides a detailed map and a flexible search of the human and mouse genome. Compared to other tools using the UCSC genome, we processed the newest GENCODE data thus can provide more information for the non-coding regions. Importantly, the runseq2gene() function is designed to find more target gene candidates for a given genomic locus using a customized search radius, which will help users to study trans-regulation. In fact, we have applied the method for the biological knowledge discovery (PLoS Genet. 2014 Oct; 10(10): e1004604, PLoS Genet. 2014 Oct; 10(10): e1004604.http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4214600/).

2) The se2pathway package implements a unique gene2pathway algorithm termed FAIME (PLoS Comput Biol. 2014 May;10(5):e1003609. http://www.ncbi.nlm.nih.gov/pubmed/22291585) and other three widely-used gene-set analysis (GSA) approaches. It also provides a way to include user-defined gene-sets, which is an adding to the current GSA tools and is important for the discovery of function. Unlike conventional GSA approaches, the package also calculates a corrected background for a more accurate Fisher's exact test (We have introduced the algorithm at BMC Medical Genomics 2015, 8(Suppl 2):S6 http://www.biomedcentral.com/1755-8794/8/S2/S6).

3) Furthermore, the package provides end-users a one-command option to find enriched pathways from 'omic' data.

Looking forwards to more users and comments and we will further improve the package.

-Holly

On 2015 Aug 21, Guangchuang Yu commented:

not runnable, see the post.

On 2016 May 11, Jonathan Heald commented:

Posted on behalf of the American Academy of Sleep Medicine.

Consensus Conference Panel., 2015

On 2016 May 05, Jonathan Heald commented:

None

On 2016 Mar 16, E M Sellers commented:

Nice - thanks for doing this ed

On 2016 Mar 13, Daniel Schwartz commented:

The CIWA-Ar scale can be used online or in a mobile application at http://qxmd.com/calculate/ciwa-ar

Conflict of interest: Medical Director, QxMD

On 2017 Jul 07, Morten Oksvold commented:

This article should have been retracted after an investigation by The University of Maryland found this article to contain "compromised" data (a total of 26 articles in 11 journals were affected). The journal Molecular Cancer Therapeutics was informed in August 2016, according to Retraction Watch.

http://retractionwatch.com/2017/04/26/university-asked-numerous-retractions-eight-months-later-three-journals-done-nothing/

On 2015 May 19, L Charles Murtaugh commented:

The basis of this paper will be somewhat confusing to those who adhere to the traditional definition of "transcription factor," as a DNA-binding protein that regulates gene transcription (http://www.nature.com/scitable/definition/general-transcription-factor-transcription-factor-167). Not one of the genes mentioned in the abstract is a transcription factor, by the conventional definition; instead, all of them are regulators of DNA synthesis or other aspects of cell cycle progression. In fact, only one of the 20 top "transcription factors" described as differentially-expressed in this study (Tables 3 and 4), Xbp1, is an actual transcription factor. I must question, therefore, the validity of the dense transcriptional regulatory networks presented in Figures 3-4 of this paper.

One culprit here seems to be the authors' uncritical acceptance of a previous study (http://www.ncbi.nlm.nih.gov/pubmed/18713790) that purported to establish a comprehensive database of mammalian transcription factors. Again, a cursory examination of this database (http://itfp.biosino.org/itfp/TFViewer/human_tf.jsp?table=human_tf&page=1) reveals that the majority of its members are not transcription factors at all. Indeed, the fact that this previous manuscript suggested the existence of >4000 human TFs, 2-3 times more than conventional estimates, should have raised red flags not only for its original peer reviewers, but also for authors of the current study.

On 2015 Jul 17, Andrea Messori commented:

Cumulative pairwise meta-analysis to evaluate the risk of infection in patients with rheumatoid arthritis treated with biological agents as compared with DMARDs: the difference is not significant using the random-effect frequentist model.

By Andrea Messori

HTA Unit, Regional Health Service, 50100 Firenze (Italy)

In the article by Singh and coworkers [1], biological agents were compared with traditional disease-modifying antirheumatic drugs (DMARDs). This comparison showed an increase in serious infections for patients treated with biological agents. To reach this conclusion, a frequentist cumulative meta-analysis (shown in Figure 4 of ref.1; 59 trials) was carried out in which the summary odds-ratio (calculated according to the Mantel-Haenszel fixed-effect [FE] model) was estimated to be 1.27 (95% confidence interval [CI]: 1.05 to 1.52).

Since Singh et al did not test one of the most commonly used meta-analytic models (i.e. the random-effect [RE] model according to Der Simonian and Laird[2]), we have re-analysed the data shown in Figure 4 of Singh’s article by running the above mentioned RE model as implemented in the Open Meta-Analyst software (version 4.16.12, Tufts University, url http://tuftscaes.org/open_meta/).

Our results are presented in the figure available at http://www.osservatorioinnovazione.net/papers/singhs-reanalysis.jpg

The odds-ratio that we obtained (1.25; 95%CI: 0.92 to 1.71; p=0.16; Figure) differs from the values reported by Singh et al. [1] mainly because “our” odds-ratio remained far from the threshold of statistical significance, whereas all the odds-ratios reported by Singh et al. satisfied this criterion.

References

[1] Singh JA, Cameron C, Noorbaloochi S, Cullis T, Tucker M, Christensen R, Ghogomu ET, Coyle D, Clifford T, Tugwell P, Wells GA. Risk of serious infection in biological treatment of patients with rheumatoid arthritis: a systematic review and meta-analysis. Lancet. 2015 May 11. pii: S0140-6736(14)61704-9. doi:10.1016/S0140-6736(14)61704-9. [Epub ahead of print]

[2] DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7:177–87.

On 2015 Nov 02, Christopher Southan commented:

Thanks to the authors for sending the 28 CID codes now mapped into PubChem http://cdsouthan.blogspot.se/2015/10/a-kinase-gift-horse.html

On 2015 Sep 08, Alexandra Alexiev commented:

This paper was featured in a microBEnet post here: http://microbe.net/2015/09/06/important-paper-on-how-biases-in-dna-extraction-can-shape-inferences-from-metagenomics-re-spore-formers/

MicroBEnet is a blog that writes about microbiology of the built environment and is funded by the Sloan Foundation to do outreach and science communication.

On 2016 Mar 22, Christopher Grefen commented:

FRET vector maps and sequences can be downloaded from my webpage.

I have prepared this little "How-to guide" for primer design of 2in1 Entry vectors; it can be accessed here.

For any other questions or vector requests, feel free to contact me via email: christopher.grefen@uni-tuebingen.de

On 2015 Aug 26, Donald Forsdyke commented:

DIMINISHED ROLE FOR CONVENTIONAL NATURAL SELECTION

This study decisively demonstrates both that nucleic acid level forces drive amino acid composition, rather than the converse, and that, in this respect, higher oligonucleotide frequencies are more powerful than mononucleotide frequencies (base composition). This is consistent with the case, made on different grounds, that oligonucleotide frequencies drive mononucleotide frequencies (summarized in Forsdyke 2011). Furthermore, there is now better support for Grantham’s “genome hypothesis” that natural selection by way of the conventional environment, may be secondary to some other form of selection that relates to speciation (see comment on Goncearenco A, 2014). Indeed, in some cases, amino acids in a protein may be mere “place holders” - there to serve the needs of the genome (Rayment JH, 2005).

Of course, some adaptation takes place at the protein level, but that the authors’ reading frame-specific analysis provides “contravening evidence” against the power of oligonucleotides is not readily apparent. In thermophiles the low frequency of TpA overlapping successive codons (e.g. NNT,ANN, …), and the depletion of ApT when positioned within a codon (e.g. NAT, NNN, … ), are easily explained by the pressure on thermophiles to purine-load their coding sequences (i.e. there is a nucleic acid level selective pressure). Thermophiles can best achieve this, without imposing excessively on amino acid composition, by incorporating purines in third codon positions. Thus, instead of the classical distribution of purines (R) and pyrimidines (Y) in codons (e.g. RNY, RNY, …), thermophiles tend to follow the RNR rule (e.g. RNR, RNR, … ) (see Lambros RJ, 2003).

Forsdyke DR: Evolutionary Bioinformatics. 2nd edition. New York: Springer, 2011.

On 2017 Jan 31, Robert Garry commented:

It is critical that the scientific literature provides an accurate account of the beginnings of the historic 2013–2016 epidemic of Ebola virus disease in West Africa. Wauquier and coauthors report incorrect information regarding the earliest known cases of Ebola in Sierra Leone. The timeline of the transmission chain that initiated the outbreak in Sierra Leone began earlier than described and the individual who performed the initial Ebola diagnosis in a Sierra Leonean is not properly acknowledged.

The date of death of a traditional healer from the village of Kpondu, Kailahun District in Sierra Leone differs from a timeline established by two independent highly credible investigations. Other publications that referred to the healer did not provide a date of her death, indicating that it occurred at an uncertain date before mid-May 2014 (1). Wauquier et al. state that, “On April 28, the healer became extremely ill and died two days later.” Investigations by the Awareness Times of Sierra Leone under the direction of Dr. Sylvia O. Blyden, who is currently Sierra Leone’s Honorable Minister of Social Welfare, Gender and Children Affairs, and separately by Dr. Sheri Fink and colleagues of the New York Times placed the date of death of the healer at or around April 8, 2014, three weeks earlier than April 30, 2014 (2, 3).

Wauquier and coauthors state that, “Ultimately, most of those who attended her [the healer’s] funeral ceremony became sick, each infected with the yet-to-be-identified Ebola virus.” This account is challenged by video taken at the funeral of the healer clearly documenting that hundreds of individuals were present (4). Only 14 confirmed Ebola cases are known to have attended the healer’s funeral. The scientific literature should reflect the fact that it is highly unlikely most individuals who attended the funeral of the healer become infected with Ebola virus.

While the first 14 cases of Ebola diagnosed in Sierra Leone are likely to have attended the healer’s funeral these individuals, all females, were diagnosed between May 25 and 31, 2014 (5). The latency period of Ebola from time of infection to symptoms is generally considered to be 2-21 days. The approximately 50 day delay to diagnosis from the actual date of the funeral circa April 8, 2014 indicates that these 14 individuals were unlikely to have been infected by contact with the healer’s corpse. Furthermore, Gire et al. (5) demonstrated that these 14 individuals were infected by Ebola virus of two genetically distinct lineages (Clade 1 and 2). This represents further strong evidence against direct infection of these individuals by pre- or post-mortem contact with the healer.

Conspicuously absent from Wauquier et al. are details of the early transmission in Sierra Leone documented in a March 19 memorandum from the World Health Organization [WHO], and emailed to a Tulane University co-author of Wacquier et al. on April 1, 2014 by a representative of Médecins Sans Frontières [MSF] (2, 3). The WHO memorandum described a probable case of Ebola who lived in or near the healer’s village. This probable early Sierra Leone Ebola case died approximately March 3, 2014. She is plausibly linked to the healer through a relative who was a close friend or assistant of the healer. Details of the March 3, 2014 death of a Sierra Leonean likely from Ebola were reported to have also been discussed in the field with the Wauquier et al. co-author by a MSF field operative.

Wauquier and coauthors state that, “On the afternoon of May 25th, the laboratory received the first blood sample from Koindu. Using reagents provided by the US Critical Reagents Program (CRP) in coordination with the US Army Medical Research Institute of Infectious Diseases (USAMRIID), the sample was analyzed by real-time reverse transcription polymerase chain reaction (RT-PCR), along with a batch of other routine samples received that same day.” The blood sample was received by Mr. Augustine Goba, Director of the Kenema Government Hospital [KGH] laboratory. Mr. Goba performed RT-PCR with detection by agarose gel electrophoresis showing that the sample sent from the Public Health Unit at Koindu was positive for Ebola virus (3). This sample was from a person living in the village of Sokoma, who attended the healer’s funeral. Confirmatory results using the USAMRIID RT-PCR were obtained only after Mr. Goba’s results had been obtained. Mr. Goba was awarded a Presidential Citation from His Excellency Ernest Bai Koroma for diagnosing this first case of Ebola in Sierra Leone.

Several co-authors of Wauquier et al. have been supported by grants and contracts on which I am the Principal Investigator. These grants and contracts contributed to the early epidemiological investigations of the Ebola outbreak in Sierra Leone. I was not informed about the submission of Wauquier et al. I also state that did not become aware of the March 19 WHO Memorandum until almost one year after April 1, 2014.

1. J. S. Schieffelin et al. (2014). Clinical Illness and Outcomes in Patients with Ebola in Sierra Leone. N Engl J Med 371: 2092.
2. K. Sack, S. Fink, P. Belluck, A. Nossiter, “How Ebola roared back,” New York Times, December 29, 2014.
3. A. Goba et al. (2016). An Outbreak of Ebola Virus Disease in the Lassa Fever Zone. J Infect Dis 214 (supplement 3):S110.
4. D. Edge, S. Achilli. (2015). Outbreak, PBS.
5. S. K. Gire et al. (2014). Genomic surveillance elucidates Ebola virus origin and transmission during the 2014 outbreak. Science 345:1369.

On 2015 Jun 03, thomas samaras commented:

Body size is another factor related to differences in male vs. female longevity. For example, I found that American males were 9% taller than females and their life expectancy at birth was 9% less. Another researcher found that when he compared men and women of the same height, their longevity was nearly the same. Rollo found that differences in longevity between male and female mice was due to their differences in body mass. Smaller male dog breeds live longer than females in larger breeds. While women and men have the same length telomeres at birth, men have shorter telomeres at older ages due to higher telomere attrition rates promoted by the creation and maintenance of trillions of additional cells during their lifetimes.

Based on many species, Moore and Promislow also independently reported that while males generally have higher mortality, when the female of a species is larger, she has a higher mortality compared to the smaller male.

Sexual longevity dimorphism is discussed in Samaras TT. Evidence from eight different types of studies showing that smaller body size is related to greater longevity. Journal of Scientific Research & Reports, 2014: 3(16): 2150-2160, 2014; article no. JSRR.2014.16.003

On 2015 Dec 28, Marco Weiergräber commented:

None

On 2015 Dec 07, Marco Weiergräber commented:

None

On 2015 Nov 10, Marco Weiergräber commented:

None

On 2015 Nov 10, Toni Schneider commented:

The comment of Marco Weiergräber is full of speculation. But scientific progress depends on careful control of novel hypotheses, especially when results of a similar research project are opposite. Scientific reports must mention opposite results, when new data are published. Siwek et al (Sleep 2014 May 1;37(5):881-92) did not refer to our results published a year earlier (Somnologie, September 2013, Volume 17, Issue 3, pp 185-192) but they speculate since then in an unscientific manner about our data, which were presented in our publication in an absolute transparent way (single data, in parallel to the resulting mean values). Our data are as reliable as the data from Siwek et al (2014). Taking this premise serious, one has to think about reasons for differences of results in an objective way. Two different mouse models were used in the two sleep studies mentioned. Logically, one must look for differences in these two mouse models, which we have discussed in an objective and fair way, without questioning the careful investigation done by Siwek et al (2014). To think about the different remnants left in the two different Cav2.3-knockout mouse lines should generate new hypotheses instead of condemning the results of a competing laboratory. We estimate the risk of an aminoterminal Cav2.3-peptide (resulting from the expression of exon 1) lower to contribute to calcium current disturbances as the risk of a "hemichannel".

The last chapter of Marco Weiergräber's comments also stays speculative, as long as he has not tested the transfer capacity of the transmitter device under discussion (a F20EET radiotransmitter from DSI). We tested the frequency bands under standardized conditions and can confirm that the bandwidth is broader than mentioned by him. Instead of repeating again and again the same critisism without mentioning e.g. a correction published by us (Schneider T. and Dibué M., 2015 in Somnologie 17, 307-308) and without presenting new proofing data, it seems to be a fight for something else but not for progress in Science.

On 2015 Oct 30, Marco Weiergräber commented:

This review article is dealing in major parts with a comparison of the work of Münch et al., "Cav2.3 E-/R-type voltage-gated calcium channels modulate sleep in mice" (Somnologie, September 2013, Volume 17, Issue 3, pp 185-192) and Siwek et al. "The Cav2.3 R-type voltage-gated Ca2+ channel in mouse sleep architecture" (Sleep. 2014 May 1;37(5):881-92. doi: 10.5665/sleep.3652). Following the publication of Münch et al. (2013), our group has raised substantial criticism to their work. For detailed information see: Weiergräber: "How do Cav2.3 voltage-gated Ca2+ channels affect sleep architecture?" (Somnologie, December 2013, Volume 17, Issue 4, pp 304-306) and Weiergräber: "Cav2.3 R-type Ca2+ channels in mouse sleep architecture—an update." (March 2015, Volume 19, Issue 1, pp 61-62) and Weiergräber: "A scientific assessment of Cav2.3 voltage-gated Ca2+ channels in rodent sleep architecture." (Sleep. 2015 Mar 1;38(3):501-2. doi: 10.5665/sleep.4520).

We had identified major experimental and analytical drawbacks in the study of Münch et al. (2013) that require -to our opinion- a total reperformance of the small sleep study performed by Münch et al. (2013). The numerous fundamental deficiencies in Münch et al. (2013) are not listet here again. We strongly advice reading our aformentioned commentaries.

However, instead of reperforming their study, Schneider and Dibue replied that the different mouse models might account for the differences in sleep study results between the groups (Schneider and Dibue, Somnologie, December 2013, Volume 17, Issue 4, pp 307-308 and Schneider and Dibue, Sleep. 2015 Mar 1;38(3):499. doi: 10.5665/sleep.4518). Although scientific reality tells a different story, Schneider and Dibue continue to suggest that the Schneider model of Cav2.3-/- might be superior to the Miller model of Cav2.3-/- which we used. In our response (Weiergräber: "A scientific assessment of Cav2.3 voltage-gated Ca2+ channels in rodent sleep architecture." Sleep. 2015 Mar 1;38(3):501-2. doi: 10.5665/sleep.4520) we have clearly elaborated that this explanation which is again repeated here in this review does not hold true and has no scientific basis. The suggestion that Cav2.3 fragments in the Miller model could form functional channels is not only speculative, indeed, publications by Annette Dolphins group (e.g. Raghib et al., 2001) has shown that it simply does not occur. Although we have communicated this fact to the authors repetitively, they again ignore it and they also did not cite our commentaries which might have helped to avoid their misleading argumentation again.

In order to comment or review their data it's mandatory for Münch et al. (2013) to first reperform their experiments to acquire valid sleep data, second to analyze for compensatory mechanisms in their model (which we did in the Miller model, but they did not in their model, at least they did not report) and finally, the Schneider model has to be evaluated for potential N-terminal peptide fragments and their influence on Ca2+ channel physiology. Finally, the authors cite Dibue et al. (2014, Epilepsia). We have explained to the authors in detail (Weiergräber and Papazoglou, Epilepsia. 2015 Jul;56(7):1180-1. doi: 10.1111/epi.13041.) that the gamma data presented there are not valid as the analysis of gamma up to 500Hz is not possible with the transmitter type they used (a F20EET from DSI with 1-50Hz bandwidth and 250 Hz nominal sampling rate). The Nyquist-Shannon-limit does not allow for analysis of such high gamma as Dibue et al. did. There is thus no justification to cite this publication. Given the misleading discussion here, a correction and clarification is requested for this part of the publication.

On 2015 May 14, Bernard Baars commented:

Jon Schooler and coworkers have conducted a wonderful series of studies on the stream of thought, distractibility, and the like. My remarks are not meant to diminish their great contribution.

Spontaneous mentation has a long and intellectually important history, including James, Freud, and Proust. I believe none of those people would believe that the spontaneous stream of thought is "mind wandering." That terminology suggests something that is dysfunctional. The alternative hypothesis is that the spontaneous stream of consciousness is indeed FUNCTIONAL, though its functions may be implicit. That is indeed the outcome of Jerome S. Singer's years of study, suggesting that "current concerns" are the focus of spontaneous thought. That includes recent difficulties in relationships, and possible solutions. It may include recurrences of traumatic memories or related cues. It may include inner complaints about the boring task, which people usually avoid if they get a chance to.

The competition between ASSIGNED tasks and spontaneous mentation may occur simply because the spontaneous stream is NEEDED and generally FUNCTIONAL. The mere fact that subjects sign a consent form does not mean they really consider the assigned task to have a higher priority compared to wondering about the attractive cabin attendant.

It is something of a scientific leap to label any preferred behavior to be dysfunctional. In clinical psychology and psychiatry such labeling only occurs after much debate, research, and review. Terms like "on-task" vs. "off-task" mentation are value neutral, and therefore preferable to "mind wandering". In one current literature on depression, sheer mind wandering is pathologized as "rumination," and taken to be a sign of depression. That would make Picasso's and Mozart's spontaneous thinking about painting and music to be dysfunctional and a sign of depression.

Not all task-independent thought is bad.

On 2015 Jul 06, Adrian Barnett commented:

I have used the random effects models on skewed exercise data with just 2 to 3 observations per person and the model converged and gave a meaningful answer. That may not always happen, but given the potential change in interpretation it is worth trying.

On 2015 May 29, Chris Wallace commented:

Thank you for your interest in our paper. We used the cosinor model because we wanted a parsimonious model for testing for seasonality. I completely agree, that to model and understand the specific seasonal patterns more precisely we will need to consider models with more parameters which do not enforce symmetry or a sinusoidal shape. Your point regarding individual random effects for the cosine and sine terms is interesting, but I wonder, does this require many repeated measures per individual to fit well? Typically the datasets we accessed, although longitudinal, had a limited number of observations per individual.

On 2015 May 16, Adrian Barnett commented:

This is an interesting and highly novel paper. The cosinor model uses only two parameters to create a seasonal pattern and is therefore parsimonious, however more complex seasonal models are likely to provide further understanding of the seasonal patterns shown.

Firstly the cosinor model used gives the seasonal pattern at a population level, but the seasonal pattern in individuals is likely to be stronger. As an extreme example, imagine half a study sample had a seasonal peak in early January and the other half in early July. At the population level these patterns would cancel out despite their being potentially strong individual seasonality. The strength of the individual seasonal pattern can be estimated by including random effects (at the individual level) for the cosine and sine terms. The greater the heterogeneity in individual seasonal patterns, the greater the difference between the individual and population seasonal pattern.

Secondly the cosinor model is symmetric in terms of both the peak and trough, and the rate of seasonal increase and decrease. More general models, such as splines or models that use a categorical variable for month, use more parameters but are able to show a wider variety of patterns, such as a steep increase during the autumn to winter transition and slower decrease during the winter to spring transition.

On 2017 May 26, Seán Turner commented:

Type strain of Sphaerisporangium corydalis is DSM 46723, not DSM 46732.

On 2015 May 18, Tony Larkman commented:

Recent advances in the detection of adulteration in tea tree oil (TTO) published by Wong et al in May 2015 (Enantiomeric distribution of selected terpenes for authenticity assessment of Australian Melaleuca alternifolia oil: http://www.sciencedirect.com/science/article/pii/S0926669015000680) show a significant number (>50%) of commercial samples varying markedly from the ∼68.5% expected value found for (+) terpinen-4-ol. It is notable that of the 15 commercially sourced samples from the European Union, 73% of these showed significant differences in chiral abundances indicating widespread incidence of this fraudulent practice.

In this article Santesteban Muruzábal et al state that the incidence of allergic contact dermatitis was "...until recently...infrequent in our setting" and "Further, three of the five patients also reacted to oxidized d-limonene". While limonene is present in pure TTO at ~1%, if the TTO is correctly stored the level of oxidation of not only the limonene but all other components remains low. More detail on this is available in a paper by Brophy et al: Gas chromatographic quality control for oil of Melaleuca terpinen-4-ol type (Australian tea tree) http://pubs.acs.org/doi/abs/10.1021/jf00089a027

In Australia, and other countries, when pure TTO steam distilled from Melaleuca alternifolia is correctly handled and stored to minimize degradation through oxidation the incidence of allergic contact dermatitis appears to be far less than that experienced in the European Union.

It is therefore not unreasonable to expect that the reactions noted may be caused by the use of adulterated material masquerading as TTO - this material is often manufactured using a poor quality (highly oxidized) product diluted with industrial waste sourced from 'normalising' other essential oils such as Pine and Eucalyptus. This industrial waste is uncontrolled and a diverse range of substances of unknown origin have also been detected when analyzed (personal observation: 2013 - 15).

It would be interesting to know if the enantiomeric ratios for the material used in this study conformed to the expected ratios as published by Wong et al.

On 2016 Aug 26, Diana Frame commented:

This article has been retracted because sections of its text are identical to another article (Afshar et al 2014, Obes Surg, PMID 25015708). The notice of retraction is featured on the publisher page, but should be shown more prominently in the PubMed record.

On 2017 Apr 12, Siegfried Hekimi commented:

We obtained very different results that suggest that CLK-1 functions exclusively in ubiquinone biosynthesis and not in the nucleus to affect the mtUPR. In a paper entitled: A single biochemical activity underlies the pleiotropy of the aging-related protein CLK-1. By Liu et al. http://www.nature.com/articles/s41598-017-00754-z?WT.feed_name=subjects_energy-metabolism

On 2015 May 15, Kausik Datta commented:

"P value of <0.1 was set as a meaningful difference with statistical significance."

I am curious, how exactly was the acceptable P-value set at a non-traditional, less robust/discriminatory value? How was this considered 'meaningful', based on... which consideration(s)? Did the peer reviewers for this article raise this question to the authors, and if so, what was the authors' response?

(Caveat: I don't have access to the full paper and, therefore, don't know if these questions have been answered in the discussion.)

On 2015 Sep 06, Lydia Maniatis commented:

The claims of Blakeslee and McCourt are flawed on logical, methodological, theoretical, and empirical grounds.

Perhaps the core error is the denial of perceptual facts, as described below (and as noted also by Gilchrist in his commentary on this article).

In the Adelson checker-shadow illusion (to take one example), a check in apparent shadow looks white, while an equiluminant check, apparently in plain view, looks black. Aside from the additional presence of the apparent shadow, the experience of the two surfaces is similar to looking at a white check and at a black check under homogeneous illumination. Kohler (1935), describes a “real-world” version wherein a white and a black paper appear white and black, respectively, even if the illumination is adjusted so that the two surfaces are actually equiluminant.

Imagine, now, what would happen if we asked a naive observer to report on “the intensity of the light coming from the surface of each of the two checks.” First, he or she would likely assume the question referred to the apparent illumination of the surfaces. In this case, the white, “shadowed” check should receive a lower rating than the black “plain-view” one. If we then tried to clarify that we want the observer to make matches based on “the amount of light each surface is sending to the eye,” (its luminance) I think the observer would have trouble a. understanding what we are asking and b. performing the required task. And I'm not sure anyone would be able to judge, with confidence, whether the two checks in the Adelson checkerboard are, or are not, in fact equiluminant. That's what makes the demo so impressive. Even if observers could estimate this value, the task would be difficult and the results unreliable. To achieve it, they would have to focus narrowly on each square, isolating it from the surround.

Yet, Blakeslee and McCourt maintain that the latter task, which requires viewers to overide their spontaneous, salient perceptual experience, is “strictly based on appearance,” while the former, effortless experience is “based on an inferential judgment.” If, however, we define “appearance” as “what something looks like,” then the authors' arguments are obviously false, as can be confirmed by any observer.

The authors' argue that the experience which is quite literally based on appearance, is actually a product of learning. This is a major claim (implying that learning can actually alter a percept from black to white), but the authors offer no evidence for it. All of the arguments and available evidence is against. As Gilchrist (2015) points out, even fish seem to naturally achieve this kind of learning. A child can see the Adelson checker-shadow effect as effortlessly as an adult. Our perceptions aren't affected by what we learn about the nature of light and the properties of surfaces, we don't have to learn how to judge when a surface is in shadow or merely darker than its neighbor, or when it is covered by various types of transparency, we don't even have to learn that at night things don't actually change color. Given the difficulty scientists have in analyzing and modelling percieved lightness, and given that massive, early exposure to artificial images mimicking illumination variation has no discernible effect on our perception of the “real” world, the claim that people go through a process of learning to make the inferences necessary to achieve veridical percepts in natural conditions does not seem credible. B and M have certainly not tested it.

The quality that the authors argue is “strictly based on appearance” is a quality that they term “brightness,” defined as the perceptual correlate of luminance. The view that there is a perceptual correlate of luminance seems to be uncontroversial among lightness researchers (e.g. Kingdom, 2011; Gilchrist, 1999)), although Anderson (2014) seems to define brightness (more properly, in my opinion) as apparent illumination. The claim seems easy to refute.

Suppose we observe a set of surfaces lacking cues to differential illumination, and that some appear white, some gray, some black. Suppose, then, that we observe the same set of surfaces, at a different time, under a different degree of illumination. Assume that have completely forgotten the previous experience with the surfaces, and are again asked to judge their white/gray/black character. Our responses will typically be similar to those we gave in the first (now forgotten) instance. In other words, even though the illumination (and consequently the luminance) of the surfaces will have changed, our responses will remain the same. If the range of luminances is complete enough, they will, in both cases, be correlated with reflectance and not with luminance. Thus, even under homogeneous illumination, we cannot say that we are perceiving luminance, or that perception is more direct than in other situations. As always, the percept is the product of a complex visual process based on luminance values and structural assumptions.

When it comes to the case of non-homogenous (apparent) illumination, the authors seem to be treating illumination boundaries as though they were directly-perceived facts serving to support “inferentially” perceived lightness judgments. They say, for example, that “when the illumination component is clearly visible” the observer can use “brightness contrast” at the boundary to infer the magnitude of the illumination. There are a number of problems with this description.

First, if a shadow is perceived – is “clearly visible” - as the cause of the luminance boundary, then viewers are perceiving a double-layer – a surface with lightness x and an apparent shadow of darkness y lying on top of it. They are not perceiving a single “brightness” value. The authors are using the term “brightness” when they actually mean luminance.

Relatedly, the illumination boundary only becomes “clearly visible” after the visual process has inferred its presence based on the luminance structure – including the relative luminances at luminance boundaries - of the image. Whether the darker side of an edge will be perceived as being similar in reflectance, but lower in illumination than its neighbor, or as darker than its neighbor due to a lower reflectance, or any other combination of possibilities, depends on the global structure of the image. Given certain conditions, even a non-existent luminance edge may produce an apparent lightness difference, as in the case of illusory surfaces. So the argument that perceiving a surface as continuing beneath a “shadow” boundary is more inferential than perceiving a particular surface as gray due to its luminance relative to other surfaces in an image is naive. If “appearance-based” means “based on luminances”, then all perception is appearance-based. If it means that surfaces are perceived based on local luminance conditions, then it never is. Local conditions do not even determine photoreceptor activity in the horseshoe crab.

As corroborating evidence, the authors point to a few references, including Blakeslee and McCourt (2008), which is supposed to prove the existence of “brightness” judgments. Their stimuli consist of the classic simultaneous contrast demo plus variations that create weak impressions of differential illumination. The “brightness” judgments are defined as those that arise when observers are instructed to focus narrowly on the targets. This is similar to applying a mask. Effectively, we are talking about the same visual process acting on a different stimulus, not about a different type of judgment. Due to the weakness of the structural cues to differential illumination in B and M's (2008) stimuli, the ability of observers to isolate the target in this way is very easy. The demand would be much more difficult, and the results surely very different, if the stimulus had been the Adelson checkerboard.

On 2015 Aug 10, Guangchuang Yu commented:

I have analyzed the data using my Bioconductor package clusterProfiler. The enriched term found by DAVID can be reproduced with more detail terms enriched. see https://github.com/GuangchuangYu/enrichment4GTEx_clusterProfiler.

On 2015 May 14, Jeff Kiefer commented:

The functional enrichment analysis performed in this paper was done with the DAVID resource Dennis G Jr, 2003, however, this resource has not been updated in nearly 5 years http://david.abcc.ncifcrf.gov/helps/update.html.

On 2015 May 27, DANIEL BROWN commented:

Unfortunately strain B2096 8B is actually Mycoplasma gallinaceum.

On 2014 Nov 25, Harri Hemila commented:

A secondary analysis of this study has been published in Hemilä H, 2013, DOI. The secondary analysis calculated that when the participants suffered from the common cold, vitamin C increased histamine PC20 level by 3.2 fold (95% CI 2.0 to 5.1). After the participants had recovered, the effect of vitamin C was smaller so that there was a significant interaction between vitamin C effect on PC20 and the presence of the common cold (P = 0.003).

On 2015 May 09, Christopher Southan commented:

Context and additional information http://cdsouthan.blogspot.se/2014/09/anti-ebola-medicinal-chemistry-time-for.html

On 2017 Jul 10, Randi Pechacek commented:

Elisabeth Bik discusses this paper in a microBEnet blog post about the microbiome of the built environment.

On 2015 Oct 06, M Felix Freshwater commented:

If other researchers obtain the same results then the quality of the data in the ACS-NSQIP database is suspect.

On 2016 Oct 11, UFRJ Neurobiology and Reproducibility Journal Club commented:

In this article, the authors demonstrate that the response to an aversively conditioned context is different between rats that have undergone conditioning individually or collectively, and suggest that this is mediated by olfaction.

Despite the interesting findings, we would like to point out that the statistical methods used for the results shown in Figures 2E and 4C do not support the conclusions made by the authors. In Figure 1B, the illustrated result was that the collectively conditioned group presented a significant decrease (p<0.01) in fear response 24 h after training when compared to animals conditioned individually. The same kind of comparison (collective vs. individual conditioning) was performed in Figure 2E, with the test performed 3.5 h after training; in this case, no statistical significance (p> 0.1) was found (even though a trend towards reduced freezing in the collectively conditioned group can be observed in the figure). Based on this, the authors conclude that “these data argue strongly that fear acquisition was similar in both groups and that neuronal mechanisms mediating fear memory consolidation and/or expression are involved in the markedly attenuated fear memory in Group animals.”

This conclusion, however, is based on a common statistical misconception: namely, to infer that, if a significant difference at a given threshold is found between two groups in one condition (i.e. collective conditioning) but not in another (i.e. individual condition), this implies that there is a significant difference between the two conditions. This statistical reasoning is erroneous, since the differences found in both conditions can be similar in magnitude (as they seem to be in this case), and the fact that one but not the other is found to be significant at an arbitrary significance threshold can be a consequence of statistical power, variability, or mere chance. As discussed by (Nieuwenhuis S, 2011) to argue in favor of a difference in magnitude between two differences found in individual comparisons, the correct statistical approach would be to test whether there is a significant interaction between the two independent variables (in this case, group and time of testing), and not simply report that an effect was statistically significant while the other it was not.

The same error occurs in Figure 4C, in which the collectively conditioned sham group presented a significant reduction compared to individually conditioned sham rats (p < 0.05) individually, while this difference between types of conditioning was not found (p > 0.1) in anosmic rats. Although the difference in this case indeed seems larger in the sham group, results for the interaction between type of conditioning and anosmia were not shown, and the post-hoc results presented cannot be used to conclude that the sense of smell is responsible for the difference between individual and collective conditioning. This does not mean that the sense of smell does not play a role, but to support this conclusion the authors should present a statistical analysis that actually tests this hypothesis.

On 2016 Oct 11, UFRJ Neurobiology and Reproducibility Journal Club commented:

None

On 2015 Jun 06, A Martinez-Arias commented:

This manuscript claims the discovery of a new kind of Embryonic Stem (ES) cell, so called ‘region specific Epiblast stem cells (rsEpiSC). A close look at the protocols, genetics and cell behaviours indicates that what is reported is likely to be an improved method for the maintenance of EpiSCs and the closely related human ES cells; it would be misleading to describe a more stable population as a new entity.

The main finding reported here is that inhibition of Wnt signalling in cultures of EpiSCs, and also human ES cells, leads to a stable pluripotent population. As briefly indicated below, this has been reported a number of times before, as has the ability of EpiSCs to integrate in postimplantation embryos, the other finding presented here as new. The observation that stabilized human ES cells can undergo a similar integration is, however, novel, though the degree to which this happens and its value will await further experiments.

It is well known that ES and EpiS cells are, for the most part, heterogeneous populations in dynamic equilibria. In the case of ES cells this can be biased towards a stable state, called ‘ground state’, by application of two inhibitors, one for MEK and another for GSK3 [1]. It is also well known that in contrast with ES cells, when EpiSCs and human ES cells are presented with high levels of Wnt signalling they differentiate [2, 3] and over the last few years a number of reports have shown that inhibition of Wnt secretion or Wnt/ß-catenin signalling increases the self renewal of EpiSCs populations [4-6]. Thus a cocktail of Activin, FGF2 together with Wnt/ß-catenin inhibitors leads to efficient derivation of EpiSCs as well as their stable culture. This is confirmed in this report without acknowledging the earlier studies –interestingly some of these studies are referred to but only as discussing Wnt signalling in pluripotency. Having repeated this observation, the authors now notice that removal of Activin from the cocktail –but not inhibition of Activin signalling, which is not tested- improved the stability of the culture; this tweak is novel and mechanistically intriguing, though it is not pursued further. Thus the main message of the work is that EpiSCs grown or derived in the presence of FGF2 and Wnt inhibitors are stable (this is no new kind of ES cell).

With regard to the ability of the cells to integrate in the posterior part of postimplantation embryos, there are also precedents showing that EpiSCs, which indeed are not able to integrate in preimplantation embryos under normal conditions, can and will integrate in postimplantation embryos [7, 8]. The point made here that ‘stable EpiSCs integrate preferentially in posterior regions of the embryo is not surprising as integration is likely to be easier in the region undergoing gastrulation, which is ongoing in the posterior region at the time of the injection. Furthermore, the conclusion that ‘rsEpiSCs’ are related to the posterior proximal epiblast of stage E6.5 is similar to that obtained by Kojima et al. that although it is possible to obtain EpiSCs from a range of stages during gastrulation, EpiSC lines tend to correspond to anterior primitive streak i.e. posterior proximal E6.5 [8]. The results presented here are compatible with the observation that EpiSCs derived from embryos at different stages of gastrulation can be different [8] i.e there might be many rsEpiSCs, which is probably not a helpful notion.

One of the arguments used by the authors to claim the identification of a new type of stem cell is their transcriptional profile. However, EpiSC lines derived from different stages around gastrulation have different properties and transcriptional profiled and much of these differences are likely to be down to signalling [8]. Thus it is not that surprising that a population of EpiSCs severely deprived of Wnt and Activin signalling exhibits a special transcriptional profile and maps, in a PCA plot, away from other cells in different states and notably from EpiSCs grown in Activin and FGF2, a very different cocktail. It is likely that under appropriate experimental conditions, rsEpiSCs will be shown to correspond to one of the EpiSCs derived by Kojima et al. [8], though the changes and adaptations associated with growth in specific signalling environments would make the comparison challenging.

In summary, this report is a protocol to obtain a state which is to EpiSCs what the ground state (2i) is to the ES cells. This is certainly useful but not enough to talk about a new kind of ES cells. Of course, this is an opinion. Nevertheless, the work provides further support to the importance of Wnt signalling in the control of the dynamics of stem cell populations.

References [1] Ying QL, Wray J, Nichols J, Batlle-Morera L, Doble B, Woodgett J, et al. The ground state of embryonic stem cell self-renewal. Nature 2008;453:519-23. [2] Singh AM, Reynolds D, Cliff T, Ohtsuka S, Mattheyses AL, Sun Y, et al. Signaling network crosstalk in human pluripotent cells: a Smad2/3-regulated switch that controls the balance between self-renewal and differentiation. Cell Stem Cell 2012;10:312-26. [3] Davidson KC, Adams AM, Goodson JM, McDonald CE, Potter JC, Berndt JD, et al. Wnt/beta-catenin signaling promotes differentiation, not self-renewal, of human embryonic stem cells and is repressed by Oct4. Proc Natl Acad Sci U S A 2012;109:4485-90. [4] Kurek D, Neagu A, Tastemel M, Tuysuz N, Lehmann J, van de Werken HJ, et al. Endogenous WNT signals mediate BMP-induced and spontaneous differentiation of epiblast stem cells and human embryonic stem cells. Stem cell reports 2015;4:114-28. [5] Kim H, Wu J, Ye S, Tai CI, Zhou X, Yan H, et al. Modulation of beta-catenin function maintains mouse epiblast stem cell and human embryonic stem cell self-renewal. Nat Commun 2013;4:2403. [6] Sumi T, Oki S, Kitajima K, Meno C. Epiblast ground state is controlled by canonical Wnt/beta-catenin signaling in the postimplantation mouse embryo and epiblast stem cells. PLoS One 2013;8:e63378. [7] Huang Y, Osorno R, Tsakiridis A, Wilson V. In Vivo differentiation potential of epiblast stem cells revealed by chimeric embryo formation. Cell Rep 2012;2:1571-8. [8] Kojima Y, Kaufman-Francis K, Studdert JB, Steiner KA, Power MD, Loebel DA, et al. The transcriptional and functional properties of mouse epiblast stem cells resemble the anterior primitive streak. Cell Stem Cell 2014;14:107-20.

On 2015 Jun 30, Prashant Sharma, MD, DM commented:

This paper basically addresses a cost-cutting issue that sometimes unfortunately still surfaces in resource-poor settings, wherein electrophoresis-derived hemoglobin fraction percentages are used for clinical decision making.

While electrophoresis followed by elution and chemical or photometric measurement is acceptable, electrophoresis followed by densitometry, we found, is simply too unreliable to be used in tests that establish genetic diagnoses or are used to screen antenatal women for fetal diseases.

We, the authors, would be happy to discuss the findings and our experiences with interested readers at the email provided at the publisher website.

On 2017 Mar 06, Atanas G. Atanasov commented:

Very interesting review dear Colleagues. I have used is as a base for writing a posting at: http://healthandscienceportal.blogspot.com/2017/03/the-health-promoting-potential-of.html

On 2016 Dec 18, Zvi Herzig commented:

The concerns for potential chronic toxicity in 2 of the 42 tested products refer to (1) terpene hydrocarbons in relation to Maximized Survey-Derived Intake (MSDI) levels and (2) diacetyl in relation to NIOSH-recommended safety limits.

MSDIs are the normal exposures considered when food safety standards are established. Exceeding them isn't necessarily hazardous, but it implies a level not considered in the assessment. Therefore, it's important to consider other available safety limits when possible.

The strictest 8 hr occupational exposure limit for terpenes is 20 ppm (111 mg/m3) Granström KM, 2010. Occupational safety limits consider a ~10 m3/workday respiration rate for active workers Kuempel ED, 2015. This corresponds to a safe terpenes inhalation limit of 1,110 mg/workday.

The highest concentration of terpenes in this study is 106.7 mg/g. A mean consumption of 3 g e-liquid daily among EC users is noted. This corresponds to terpene consumption of 320.1 mg/day for the highest terpene concentration e-liquid -- a fraction of the aforementioned safety limit.

Furthermore, with a high prevalence of switching between flavors Farsalinos KE, 2013, these quantities are further diluted in relation to chronic use. This study detected hydrocarbons at 2.5 mg/g on average, all hydrocarbons being terpenes. In contrast, a single cigarette delivers 4.9 mg of hydrocarbons, including 1.1 mg of terpenes Perfett TA, 2014. Despite the much higher exposure to terpenes in cigarette smoking, terpenes are not noted in any of the lists of cigarette smoke's notable toxicants.

With regards diacetyl estimated to result in 126 µg/day exposure in one case, it should be noted that tobacco smokers inhale 5700 µg diacetyl per day Farsalinos KE, 2015, 45 times greater than the exposure resulting from using the product with the highest diacetyl level among 42 liquids tested in this study.

On 2016 Apr 05, Marko Premzl commented:

The third party data gene data sets of eutherian adenohypophysis cystine-knot genes (128 complete coding sequences) and D-dopachrome tautomerase and macrophage migration inhibitory factor genes (30 complete coding sequences) HF564658-HF564815 were deposited in European Nucleotide Archive under research project "Comparative genomic analysis of eutherian genes" (https://www.ebi.ac.uk/ena/data/view/HF564658-HF564815). The 158 complete coding sequences were curated using tests of reliability of eutherian public genomic sequences included in eutherian comparative genomic analysis protocol including gene annotations, phylogenetic analysis and protein molecular evolution analysis (RRID:SCR_014401).

Project leader: Marko Premzl PhD, ANU Alumni, 4 Kninski trg Sq., Zagreb, Croatia

E-mail address: Marko.Premzl@alumni.anu.edu.au

Internet: https://www.ncbi.nlm.nih.gov/myncbi/mpremzl/cv/130205/

On 2018 Jan 19, Natalie Clairoux commented:

Free version of this article available at http://hdl.handle.net/1866/19737

On 2016 Sep 19, John Marshall commented:

None

On 2016 Sep 17, Christopher Southan commented:

The non-disclosure of the Genz-682452 molecular structure (even though it may have been exemplified in WO2012129084) renders this work irreproducible

On 2017 Mar 19, NLM Trainees’ Data Science Journal Club commented:

The NLM Trainees Data Science Journal Club for this discussion consisted of several NLM staff members, including one of the paper’s authors. In brief, the authors employed a visual and a computational method to explore the contribution of individual drugs to a drug class-level signal in the context of adverse events. The authors used both techniques to identify class effects caused by all members of a drug class individually, as well as class effects based on a subset of drug class members.

As background information regarding adverse events, one group member pointed out that there has to be enough evidence between a drug or class and an adverse event for it to be formally labeled an adverse event. As such, the type of comparison performed in this study has the potential to introduce false positives, where there is a co-occurrence between a drug and an event that is not to the level of a “label-able” event.

In terms of methodology, the group discussed that, while this study is biased towards using case reports, the strength of evidence would likely be higher if based on randomized controlled trials; however, RCT’s are generally more focused on drug efficacy than safety.

The question was raised as to whether study results might differ if the drugs or adverse events were mapped using different terminologies than ATC and MeSH, respectively. In response, the author pointed out that the FDA uses MEDRA to report events, but for the purposes of this study, it was more complicated to use than MeSH, as it would have required hand curation (beyond the automated curation the UMLS offers) to be useable. That said, the general methodology used by the authors is largely terminology agnostic and can be used with any terminology that is reasonably hierarchical.

One group member was curious whether this process could be used for clinical decision support. Discussion revealed that this methodology is not suited for CDS, as the evidence would need to be clearer. For CDS use, the process would need to be based on drug labels or drug information systems, which are usually expressed at the individual drug level. It might be interesting to know if a class was affecting adverse events at the class level or based on a subset of the class, but this would not be the primary concern in a primary care setting. However, the results of the study could be valuable to drug safety professionals and those building adverse event repositories to support their review or decisions in regard to adverse event relationships.

On 2016 May 18, Heidi Schulz commented:

The c.910_912delGAT mutation was already reported in Lotery et al. 2000 (PMID10798642), Kinnick et al. 2011 (PMID21273940) and Katagiri et al. 2015 (PMID26201355).

On 2017 Jul 01, L-T Jia commented:

the additional file (https://reproductive-health-journal.biomedcentral.com/articles/10.1186/s12978-015-0031-x) of the study seemed unrelative with morphokinetic illustration of the embryo development.

On 2016 Aug 01, Joaquim Radua commented:

Re: the first comment, I think there may be some unfortunate confusion. Raw p-values of current voxelwise meta-analyses have not the same meaning as usual p-values because they are not derived from the usual null hypothesis (“there are no differences between groups”), but from another null hypothesis (“all voxels show the same difference between groups”). Thus, up to the moment one of the only ways to "approximately" know if the results of a voxelwise meta-analysis are neither too liberal nor too conservative is to compare them with the results of a mega-analysis of the same data, and that's what it was done.

On 2016 Jul 14, Lars Schulze commented:

Dear Christopher Tench,

we agree that it is important to control the rate of false positives in meta-analyses. Please note, that our study did apply empirically validated thresholding procedures that have been shown to not only balance sensitivity and specificity, but also to be equivalent to a corrected P-value of 0.05 (Radua et al., 2012). To even further reduce the possibility of false positive results, we used an additional voxel-wise threshold with Z-values >1.

Thus, our meta-analysis applied validated and recommended methods to control the inflation of false positive results.

Sincerely, Lars Schulze

On 2016 Jun 19, Christopher Tench commented:

The methods employed do not constitute a meta analysis as uncorrected p-values offer no protection against false positive results. Consequently the study provides no evidence of significant consistency across studies. Uncorrected p-values are not useful in meta analysis of neuroimaging studies.

On 2015 May 04, Andrew Brown commented:

This is our letter about concerns with this paper Hernández-Cordero S, 2014; the authors' response is here http://jn.nutrition.org/content/145/5/1029 (not yet indexed in PubMed).

On 2015 May 05, James Tsung commented:

Link to Case Videos: https://youtu.be/vTyiofq-UsI?list=PLHg2xyua_KjtriEZcFW6tH0iIIPTwSQvT

On 2017 Jul 21, Horacio Rivera commented:

Are citations related to a faulty writing style or simply to content? The results got by Weinberger et al. (2015) in their analysis of >1,000,000 abstracts may not be so surprising. Taking into account the pervasiveness of an improper writing style (Woodford FP, ed. Scientific Writing for Graduate Students, CBE 1986; Gopen&Swan, Am Sci 1990, 78: 550; Knight J, Nature 2003, 423: 376; Editorial, Nat Struct Mol Biol 2010, 17: 139), I guess that the decreased citation rate associated with half of 15 well-known stylistic features can equally be ascribed to the papers’ content. Actually, the opposite pattern found for the remaining features suggests that style alone is not a consistent predictor of citation rates. A more reliable study would compare a sample of "well written" vs "poorly written" abstracts, classified according to conventional wisdom by expert colleagues, instead of analyzing all available abstracts in a given period. I do two additional comments: 1. Rule 4 of the authors "Use the present tense" is unsuitable when dealing only with abstracts. At least in biomedicine, the general advice is to use the past tense in the "M&M" and "Results" sections that account for half or more of an abstract. Obviously, the results would have been the opposite ones if the rule "Use the past tense" had been applied. 2. The assertion that in "writing a paper, the limiting step is the ability to find the right article" appears nonsense inasmuch it disregards the crucial intellectual input required for successful writing. In other words, "to find the right article" is a preparatory rather than limiting step. To conclude, I praise the effort of Weinberger et al. (2015) in summarizing ten simple rules aimed to improve our writing style.

On 2016 Oct 02, Atanas G. Atanasov commented:

Readers might also find of interest the study “Does a Graphical Abstract Bring More Visibility to Your Paper?”: https://www.ncbi.nlm.nih.gov/pubmed/27649137

On 2015 May 17, Egon Willighagen commented:

Like Open Access, "sharing" is a meaningless term if it is not linked to meaningful rights. The problems outlined in this paper result from the fact that their may be a wish to share data but only if it allows you to take back the data. Private, custom data licenses do just that. There is nothing wrong with this kind of sharing, but it must not be confused with Open Data. It must not be confounded with terms like "publicly available", because if it needs a signature, it's not publicly available. That makes the lead of this article quite misleading.

For public or open data, three basic rights are part of the social agreement between the data owner (yes, fact in many countries; database rights, etc) and data user. These rights are: 1. make a copy, 2. make modifications, and 3. reshare (under the same conditions). By using a license (or waiver) that gives this rights automatically to the receiver, then there is no need for signatures. It also allows for anyone to make the mappings that are required to convert one format into another.

On 2016 Aug 23, Ben Goldacre commented:

This trial has the wrong trial registry ID associated with it on PubMed: both in the XML on PubMed, and in the originating journal article. The ID given is NCT0033633. We believe the correct ID, which we have found by hand searching, is NCT00336336.

This comment is being posted as part of the OpenTrials.net project^[1] , an open database threading together all publicly accessible documents and data on each trial, globally. In the course of creating the database, and matching documents and data sources about trials from different locations, we have identified various anomalies in datasets such as PubMed, and in published papers. Alongside documenting the prevalence of problems, we are also attempting to correct these errors and anomalies wherever possible, by feeding back to the originators. We have corrected this data in the OpenTrials.net database; we hope that this trial’s text and metadata can also be corrected at source, in PubMed and in the accompanying paper.

Many thanks,

Jessica Fleminger, Ben Goldacre*

[1] Goldacre, B., Gray, J., 2016. OpenTrials: towards a collaborative open database of all available information on all clinical trials. Trials 17. doi:10.1186/s13063-016-1290-8 PMID: 27056367

* Dr Ben Goldacre BA MA MSc MBBS MRCPsych<br> Senior Clinical Research Fellow<br> ben.goldacre@phc.ox.ac.uk<br> www.ebmDataLab.net<br> Centre for Evidence Based Medicine<br> Department of Primary Care Health Sciences<br> University of Oxford<br> Radcliffe Observatory Quarter<br> Woodstock Road<br> Oxford OX2 6GG

On 2015 May 01, David Mage commented:

"Close, but no cigar!" Human 5HT2A is autosomal. OMIM shows 182135 - 5-@HYDROXYTRYPTAMINE RECEPTOR 2A; HTR2A Cytogenetic locations: 13q14.2; SIDS and all respiratory deaths under 5 years of age have a 50% male excess suggesting an unidentified X-linked gene with a recessive non-protective allele with frequency q = 2/3 in Hardy-Weinberg Equilibrium may be involved (NB: 5HT2C is X-linked). See Mage DT, Donner EM. An explanation for the 25% male excess mortality for all children under 5. Scandinavian Journal of Forensic Science 2015;21(1) doi: 10.1515/sjfs-2015-0001

On 2016 Aug 23, Ben Goldacre commented:

This trial has the wrong trial registry ID associated with it on PubMed: both in the XML on PubMed, and in the originating journal article. The ID given is NCT0102455. We believe the correct ID, which we have found by hand searching, is NCT01024595.

This comment is being posted as part of the OpenTrials.net project^[1] , an open database threading together all publicly accessible documents and data on each trial, globally. In the course of creating the database, and matching documents and data sources about trials from different locations, we have identified various anomalies in datasets such as PubMed, and in published papers. Alongside documenting the prevalence of problems, we are also attempting to correct these errors and anomalies wherever possible, by feeding back to the originators. We have corrected this data in the OpenTrials.net database; we hope that this trial’s text and metadata can also be corrected at source, in PubMed and in the accompanying paper.

Many thanks,

Jessica Fleminger, Ben Goldacre*

[1] Goldacre, B., Gray, J., 2016. OpenTrials: towards a collaborative open database of all available information on all clinical trials. Trials 17. doi:10.1186/s13063-016-1290-8 PMID: 27056367

* Dr Ben Goldacre BA MA MSc MBBS MRCPsych<br> Senior Clinical Research Fellow<br> ben.goldacre@phc.ox.ac.uk<br> www.ebmDataLab.net<br> Centre for Evidence Based Medicine<br> Department of Primary Care Health Sciences<br> University of Oxford<br> Radcliffe Observatory Quarter<br> Woodstock Road<br> Oxford OX2 6GG

On 2017 May 29, Rashmi Das commented:

We thank Harri for his comment. It is really sad to see that Harri is trying to disseminate his personal opinion (HANDLE) THOUGH A DETAILED PEER-REVIEWED REPORT IS AVAILABLE AT COCHRANE LIBRARY. We need to share the sequence of events here, so that, things can become more transparent. After our publication of the Cochrane review in 2011, Harri published another review on the same topic in Open Respiratory Medicine Journal in same year. He provided comment on our 2011 Cochrane review in 2012, which we incorporated in 2013 version of the Cochrane review. In 2015, Harri made a report on the Cochrane review 2013 version (not 2011 version, which is still available there) and complained to the Cochrane Editor (Whatever has been mentioned in the ABOVE HANDLE: http://hdl.handle.net/10138/153180, which is a PERSONAL OPINION AND NOT A PEER-REVIEWED CONTENT). The Cochrane editors made a detailed inquiry (INVOLVED AN INDEPENDENT STATISTICIAN) and made the recommendation to withdraw the review NOT BASED ON TEXT PLAGIARISM BUT BASED ON INCONCLUSIVE DATA ANALYSIS (Detail available at: http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD001364.pub5/abstract;jsessionid=FCA801377C71E663B0004B9BBB92A908.f02t03). This report also mentions that we have acknowledged and cited his review (Hemilla 2011). But the irony is that the ABOVE HANDLE (Which contains DIRECT COPY OF THREE IMAGES/FIGURES FROM OUR PUBLICATION) IS BEING DISSEMINATED AS A PUBLICATION/REFERENCE TO SENT THE MESSAGE THAT WHAT HE HAS MENTIONED IS ENTIRELY RIGHT (BUT THE COCHRANE REPORT DOES NOT FULLY AGREE TO THIS). MOREOVER THE HANDLE IS STILL ACTIVE AFTER THE COCHRANE LIBRARY PUBLISHED THE REPORT AND REACHED THE CONCLUSION.

On 2017 May 24, Harri Hemila commented:

Background of the withdrawal

Concerns were expressed about unattributed copying of text and data, and about numerous other problems in the Cochrane review “Zinc for the Common Cold” by Singh M, 2013.

Details of the concerns are available at: http://hdl.handle.net/10138/153180.

On 2015 Jul 14, Andrew Kewley commented:

The authors show that questionnaires and other diagnostic methods lead to statistical overlap (mush) and therefore conclude that this group of disorders can be re-branded as a spectrum called "ALPIM".

The sample group of this study is highly biased and cannot be generalised, so the question is, should this study be replicated with a community/population based study?

But in the big picture, what is there to gain from this?

Medicine demands specific criteria so that medical or psycho-social interventions can be targeted to those whom will benefit.

Yes, diagnostic measure and criteria for the underlying illnesses needs to become more specific to become more useful. But ALPIM does not offer specificity of diagnosis, nor specific treatments, therefore I don't really see this as a worthwhile direction.

On 2015 May 28, Peter Good commented:

Cochran et al. measured brain glutamate, glutamine, GABA, and other metabolites by MRS in the anterior cingulate cortex of ASD adolescents to test whether autistic behavior arises from imbalance of excitatory vs. inhibitory neurotransmitters. They found normal concentrations of excitatory transmitter glutamate, but significantly high non-transmitter glutamine and significantly low inhibitory GABA. They concluded high brain glutamine and/or low GABA might explain imbalance of excitatory vs. inhibitory transmission – i.e. autistic behavior.

True, low brain GABA could be responsible – but implicating high brain glutamine in autistic behavior is risky at best, and potentially dangerous. Not only do ASD children consistently have low plasma glutamine, high brain glutamine appears to protect against ASD. This is most obvious in children with high brain glutamine from inborn urea cycle disorders (UCD) or propionic acidemia (PA) [Good 2014]. In UCD, failure of the liver urea cycle to detoxify blood ammonia at first pass allows high levels into the brain, which astrocytes detoxify by combining with glutamate to form glutamine. Krivitzky et al.: “Children in this cohort [UCD] show other behavioral/emotional strengths, including a minimal percentage with previous diagnoses of Autism spectrum disorders, mood disorders, and other psychiatric disorders.”[Krivitzky L, 2009] Saul Brusilow (MD) never mentioned autism or autistic behavior in any paper on UCD or hepatic encephalopathy (HE) [personal communication 2013]. Gropman et al., however, noted patients with partial deficiencies of urea cycle enzymes and late-onset presentations may show signs of autism [Gropman AL, 2007]. Cochran et al. also found the osmolyte myoinositol normal in ASD brains; when astrocyte glutamine is high in urea cycle disorders or HE, myoinositol is consistently low. Gabis et al., however, found myoinositol high in ASD astrocytes [Gabis L, 2008].

Propionic acidemia is another inborn metabolic disorder where high blood ammonia becomes high brain glutamine. Al Owain et al.: “In the consensus conference about diagnosis and management of PA hosted in Washington, D.C. in January 2011, there was no reported association among the neurological sequelae of the disease between PA and autism.”[Al-Owain M, 2013] Other physicians who treat PA and UCD children also report they rarely show autistic behavior. Sabine Scholl-Bürgi (MD): “In our PA patient group none has an ASD.”[personal communication 2014]. Professor of pediatrics (MD): “I see lots of kids with PA and UCD but few (perhaps none) have ASD.”[personal communication 2013]

Moreover, plasma concentrations of glutamine and GABA in ASD children are consistently opposite brain concentrations detected by Cochran et al. Plasma glutamine is consistently low [e.g. Moreno-Fuenmayor H, 1996; Aldred S, 2003; Shimmura C, 2011] and plasma GABA high [Cohen 2000; Dhossche D, 2002]. Ghanizadeh concluded: “The low level of plasma glutamine . . . is suggested as a screening test for detecting autism in children especially those with normal IQ. The decreased level has been reported before in all children with autism.”[Ghanizadeh A, 2013] Dhossche et al. thought high plasma GABA explained mood disorders and stupor. Burrus concluded a reaction between ammonia and propionic acid could produce a molecule structurally very similar to GABA [Burrus CJ, 2012].

Glutamine is normally the most abundant amino acid in blood [Souba WW, 1991], a primary brain osmolyte, alternative fuel for brain neurons and astrocytes (especially during hypoglycemia) [Stelmashook EV, 2011], and primary fuel in rapidly replicating cells (e.g. blood vessel endothelial cells, intestinal enterocytes, liver cells, and lymphocytes) [Souba WW, 1987; Souba WW, 1991; Deutz NE, 2008]. Glutamine released from skeletal muscles for anabolic responses to infection may explain the dramatic ability of fever to relieve autistic behavior [Good P, 2013]. A new clue to this phenomenon was recently published at <www.autismstudies.net>.

Autism Research Institute practitioners commonly give ASD patients oral glutamine to heal their intestines, from 250mg–8g/day, with few side effects (some hyperactivity) [Good P, 2013] – although one neurologist reported seizures. Only two practitioners, however, reported improved behavior from glutamine. Franco Verzella (MD) in Bologna, Italy gives ASD children 5–7g/day of oral glutamine after cleansing their intestines of pathogens like bacteria and Candida: “Multifactorial and multisystemic is the condition, so that the improvement has different aspects in different children. Most common: sedation, less stereotypes, better sleep, more concentration.”[personal communication 2013]

Cochran and colleagues need to think twice about reducing brain glutamine in ASD children – whose plasma and brain glutamine are more likely TOO LOW than too high. Pangborn (2013) recommended the free amino acid taurine as a natural way to increase conversion of ammonia + glutamate to glutamine.

Peter Good Autism Studies La Pine OR www.autismstudies.net autismstudies1@gmail.com

Aldred S, Moore KM, Fitzgerald M, Waring RH. Plasma amino acid levels in children with autism and their families. J Autism Dev Disord 2003;33:93–97.

Al-Owain M, Kaya N, Al-Shamrani H, et al. Autism spectrum disorder in a child with propionic acidemia. JIMD Rep 2013;7:63–66.

Burrus CJ. A biochemical rationale for the interaction between gastrointestinal yeast and autism. Med Hypotheses 2012;79:784–785.

Cohen BI. Infantile autism and the liver: a possible connection. Autism 2000;4:441–442.

Deutz NEP. The 2007 ESPEN Sir David Cuthbertson Lecture: amino acids between and within organs. The glutamate-glutamine-citrulline-arginine pathway. Clin Nutr 2008;27(3):321–327.

Dhossche D, Applegate H, Abraham A, et al. Elevated plasma gamma-aminobutyric acid (GABA) levels in autistic youngsters: stimulus for a GABA hypothesis of autism. Med Sci Monit 2002;8:PR1–PR6.

Gabis L, Wei Huang, Azizian A, et al. 1H-magnetic resonance spectroscopy markers of cognitive and language ability in clinical subtypes of autism spectrum disorders. J Child Neurol 2008;23(7):766–774.

Ghanizadeh A. Increased glutamate and homocysteine and decreased glutamine levels in autism: a review and strategies for future studies of amino acids in autism. Dis Markers 2013;35:281–286.

Good P. Does infectious fever relieve autistic behavior by releasing glutamine from skeletal muscles as provisional fuel? Med Hypotheses 2013;80:1–12.

Good P. Why do children with propionic acidemia or urea cycle disorders rarely show autistic behavior? Autism–Open Access 2014;4:3.

Gropman AL, Summar M, Leonard JV. Neurological implications of urea cycle disorders. J Inherit Metab Dis 2007;30:865–879.

Krivitzky L, Babikian T, Lee HS, et al. Intellectual, adaptive, and behavioral functioning in children with urea cycle disorders. Pediatr Res 2009;66:96–101.

Moreno-Fuenmayor H, Borjas L, Arrieta A, et al. Plasma excitatory amino acids in autism. Invest Clin 1996;37:113–128.

Pangborn JB. Nutritional Supplement Use for Autism Spectrum Disorder. San Diego: Autism Research Institute; 2013.

Shimmura C, Suda S, Tsuchiya KJ, et al. Alteration of plasma glutamate and glutamine levels in children with high functioning autism. PLoS One 2011;6:e25340–e25346.

Souba WW. Interorgan ammonia metabolism in health and disease: a surgeon’s view. J Parenter Enteral Nutr 1987;11:569–579.

Souba WW. Glutamine: a key substrate for the splanchnic bed. Ann Rev Nutr 1991;11:285–308.

Stelmashook EV, Isaev NK, Lozier ER, et al. Role of glutamine in neuronal survival and death during brain ischemia and hypoglycemia. Int J Neurosci 2011;121:415–422.

On 2015 Nov 12, RUBEN ABAGYAN commented:

Additional disclosure. One of the co-authors, R.A., has an equity interest in Molsoft, LLC. The terms of this arrangement have been reviewed and approved by the University of California, San Diego in accordance with its conflict of interest policies.

On 2016 Feb 16, Rafa Carretero commented:

Why should I publish? I wonder that question again and again, and to be honest, I don't have any suitable answer yet. It's been a long time since I decided to make a PhD and to begin to publish scientific papers. I don't know if it's worth it. I've found this article useful because I firmly believe it's the librarian who encourages me constantly to catch up with my duties of publishing, although sometimes I cannot see the point of why I should invest as much effort in publishing, rather than procrastinating. The first aim of scientific publication is to build and share knowledge. Above this romantic, idealistic thinking, it must be clear to us our true motivations to publish, because over it, we have personal motivations. It's beneficial to us. Such a very good article made me to write more on this subject, so I eventually wrote a post to add a few thoughts to those mentioned in this paper: http://www.rafalinux.com/thesis/articles/25/why-should-i-publish Thank you.

On 2015 Sep 17, BSH Cancer Screening, Help-Seeking and Prevention Journal Club commented:

The HBRC Journal Club reviewed this paper, which investigates the effects of a decision aid administered in primary care on men’s intention to undergo prostate cancer screening. The authors conducted a randomised trial in a large number of general practices in France, in a mix of rural and urban areas. The topic intersects with several challenging and controversial areas of public health policy.

The authors report that the intervention (consisting of a double-sided sheet of A4 describing various aspects of PSA testing) reduced the proportion of recruited men who intended to be screened for prostate cancer. However, although the study is stated to be a pragmatic trial, it is unclear how the intervention and associated findings apply to standard practice. Specifically, men appear to have been provided with the decision aid despite having given no indication that they were interested in PSA screening or concerned about prostate cancer, whereas these are the scenarios in which a decision aid is more likely to be used (notwithstanding possible differences between general practice in the UK and France). It is likely that men who spontaneously raise the issue of PSA screening with their general practitioner have preconceptions about the harms and benefits that are different to men who are not yet aware of it, which would lead them to respond to the decision aid quite differently. Moreover, the control condition, in which participants received only a questionnaire about prostate screening, may have had the paradoxical effect of increasing intentions among men who had not previously been considering prostate screening by raising the option without giving any indication of the associated harms.

We also discussed characteristics of the decision aid itself. Inevitably, designing information for lay people involves a balance between competing priorities: it needs to be concise enough to be manageable within a given context (general practice, in this case) but comprehensive enough to convey all the desired detail. It also needs to be comprehensible to the target audience, most of whom will not be familiar with medical terminology or statistics. These competing trade-offs are made clear in this study, in which it appears that brevity has been prioritised over completeness. For example, we noted that there is no explicit information about the (limited) benefits of PSA screening on the first page of the decision aid, nor is there an estimate of the risk of false positives and overdiagnoses within the icon array. One might take the principled view that it is important to provide this information in order for men to be able to make an informed choice. Alternatively, one might take a pragmatic view that such information would burden participants with unnecessary detail. However, what constitutes the optimal method is subject to debate and is often a matter of personal opinion.

A strength of the design was the follow-up question on the reasons underpinning participants’ screening intentions since this highlighted some of the specific viewpoints that were altered by the intervention. However, the study was limited by omitting a measure of knowledge, which is a necessary component of informed decision-making. For example, the study could have assessed whether the intervention was successful in terms of making participants aware of the possibility of overdiagnosis and lack of evidence to support a reduction in all-cause mortality.

The HBRC Journal Club found this to be an interesting study, which benefited from its large sample size, randomised design and ‘intention reasons’ as a secondary outcome measure. However, it was limited by unclear applicability to standard practice and lack of additional secondary outcomes. In particular, this study illustrates the challenges of generating universally accepted information about screening.

Conflicts of interest. We report no conflict of interests and note that the comments produced by the group are collective and not the opinion of any one individual.

On 2015 Aug 24, Eiko Fried commented:

We have published a commentary on this article in Frontiers of Psychiatry, in which we introduce Symptomics as a novel research paradigm for psychiatry and clinical psychology. We provide a brief summary of the commentary below.

Summary:

"Research has now shown that distinct depression symptoms differ in the risk factors that predispose them, their underlying biology, their response to specific life events, and their impact on impairment of psychosocial functioning. The recently published work by Hieronymus et al. adds the differential reactivity of depression symptoms to antidepressant medication to this prior body of work. The authors argue that the findings stress the importance of analyzing individual depression symptoms in future studies. We would like to extend their claim: these results mandate the examination of symptom-specific effects throughout the realm of psychopathology. Symptomics invites the application of new modeling efforts to the level of individual symptoms as fundamental building blocks of mental disorders. Focusing (A) on the level of symptoms and (B) analyzing the causal relations among them—as an alternative approach to the dominant focus on diagnoses—is likely to extend our understanding of psychopathology directly and significantly. As such, symptomics may herald a time of renewed research energy that could, finally, provide an inroad to achieve real understanding of the mechanisms underlying psychopathology."

Reference:

Fried EI, Boschloo L, van Borkulo CD, Schoevers RA, Romeijn J-W, Wichers MC, de Jonge P, Nesse RM, Tuerlinckx F, & Borsboom D (2015). Commentary: "Consistent superiority of selective serotonin reuptake inhibitors over placebo in reducing depressed mood in patients with major depression", Frontiers in Psychiatry 6, 1–3. DOI: 10.3389/fpsyt.2015.00117.

URL: http://journal.frontiersin.org/article/10.3389/fpsyt.2015.00117/full

On 2015 Sep 16, Geriatric Medicine Journal Club commented:

This is an importnat trial fo HZ/su, a a subunit vaccine containing varicella-zoster virus glycoprotein E and the AS01B adjuvant system. Post-herpetic neuralgic can be a disabling event in the elderly. This article was critically appraised at the August 2015 Geriatric Medicine Journal Club (follow #GeriMedJC on Twitter). One of the study authors was also present for the live discussion. The full discussion can be found at: http://gerimedjc.blogspot.com/2015/09/study-author-joins-us-for-successful.html?spref=tw This trial shows promise particularly in the older old where Zostavax has waning efficacy. Approval, price, and coverage remain to be determined.

On 2017 Aug 25, Lance Turtle commented:

This review states that it addresses the safety and immunogenicity of three vaccines manufactured in China, two inactivated (one PHK derived, the other Vero cell derived) and the live attenuated JE vaccine SA14-14-2. However, table 2 references Feroldi et al 2012 (PMID: 22777096), which is a study exclusively of the chimeric JE vaccine (Imojev) and does not include live vaccine SA14-14-2.

The references appear to be mislabelled which may account for some of the confusion, e.g. the section entitled “safety of the three vaccines” refers to the same paper by Feroldi et al (using a different number) as a source of data for the inactivated vaccine safety analysis.

Although the discrepancy might be accounted for purely by mis-referencing, table 2 refers to two papers by Feroldi et al, only one of which contains any data on the vaccines under review. Therefore, I suspect that the analysis of data on the live vaccine SA14-14-2 has been contaminated by inclusion of data on another vaccine not specified in the authors’ review strategy. The usefulness of this review is questionable unless these discrepancies can be corrected, and the paper re-referenced (as far as I can see all of the references in the tables are wrong).

On 2015 Jul 22, Bill Cayley commented:

A good example of when the "less-medical" approach might be better! https://lessismoreebm.wordpress.com/2015/07/22/chewing-gum-for-postoperative-recovery-of-gastrointestinal-function/

On 2016 Mar 12, Kjartan Vaarbakken commented:

Thanks for your enourmous effort undertaking this review on laser therapy for knee osteoarthritis (OA).

In an attempt to add some to the discussion by looking for dose-response relationships, I suggest that laser with continuous output of 100 to 800 mW, energy density of 48 to 50 J/cm2, and energy per point of 6 J - all performed 2 times a week - may be the thing to try in the next laser trial on knee OA? Maybe the World Association of Laser Therapy guideline has underestimated the energy needed to effectively treat an osteoarthritic knee joint? Or maybe there is deeper tissues within the joint being reached by putting in more light energy? As cartilage and bone? Or, the fact that higher laser doses is needed for anti-inflammatory effects, a high energy deliverance may be the right stimuli to impair a synovial inflammation?

Please take a look at this link supposed to illustrate a possible dose-respons relationship: https://twitter.com/kvaarbakken/status/708593031998533634

Best regards, Kjartan Vaarbakken Physiotherapist, MSc, PhD student Norwegian University of Science and Technology Trondheim, Norway http://www.ntnu.edu/employees/kjartan.varbakken

On 2016 Jun 24, Maddalena Boccia commented:

On the behalf of all the authors, I must report that the paper refers to a study performed when no bug has yet been reported in GingerAle 2.1.1 and has been published online on Aprile 2015 (accepted on April 2nd 2015), well before the bug description. Thus, there is no way to avoid this referred bug. However, taking in to account the comment posted on Pubmed 1 week ago (we are sorry to see it only now, but we haven’t received any notification about this) about results showed in the present publication, we re-run the analyses with the GingerAle 2.3.6 version, using a cluster-level inference, that it is consider the most appropriate method for statistical inference in ALE meta-analysis at this moment (Eickhoff et al., In Press “Behavior, sensitivity, and power of activation likelihood estimation characterized by massive empirical simulation” NeuroImage). We find that the main corpus of results did not change and our main conclusion, that is that specific neural networks of brain areas underpin PTSD after different traumatic events and that these networks may be related to specific aspects of the traumatic events is undoubtedly valid. In order to clear any doubt, we are considering to report the results with the last GingerAle version in a Letter to Editor, a form of scientific communication which allows to properly comment any scientific doubt raised by a paper in a polite and data-founded/evidence-based way

On 2016 Jun 14, Christopher Tench commented:

The version of GingerALE (2.1.1) has a bug that results in false positive results. The results in this publication are not valid. There was a fix at version 2.3.3

On 2015 Sep 28, Gauthier Bouche commented:

The placebo-controlled randomized trial reported in this article explores the effects of anti-inflammatory drugs (steroidal and non-steroidal) on the inflammation induced by surgery of rectal cancer. Knowing the critical role that the peri-operative period may have on long-term cancer outcomes (Horowitz M, 2015), results of this trial are of major importance for clinicians and researchers interested in this topic. Since the full article is only available in Chinese, it is important to report the number of recurrences in each of the group as only the final p-value is reported in the abstract. The number of recurrence was 1, 3 and 3 in group A (celecoxib), group B (methylpredinosolone) and group C (placebo) respectively. The statistical test used was a Chi-square test. In the method section, the authors do not specify any statistical hypothesis (power calculation) for the recurrence endpoint. Therefore the conclusion of the author that celecoxib does not lower postoperative recurrence rate of rectal cancer is not supported by the results. The trial was not sufficiently powered to detect a difference (even a large one).

On 2015 May 20, James C Coyne commented:

The registration of this trial http://www.isrctn.com/ISRCTN26666654 clearly states the hypothesis that mindfulness-based cognitive therapy (MBCT) with support to taper/discontinue antidepressant medication is superior to maintenance antidepressant medication in preventing depression over 24 months. No rationale is given for anticipating the superiority of MBCT, but it clearly was not achieved in the present study in terms of any primary or secondary outcomes. Nonetheless, this article proclaims the finding of no differences as an achievement. The study has been widely publicized in the media as demonstrating that there are no differences between MBCT and antidepressants, which is clearly not a conclusion justified by the design or the results.

The claims of the present article require a noninferiority design which requires, in turn, a much larger sample size.

In a blog post http://blogs.plos.org/mindthebrain/2015/05/20/is-mindfulness-based-therapy-ready-for-rollout-to-prevent-relapse-and-recurrence-in-depression/, I detail a number of concerns about this study and suggest it does little to fill the gaps in what little we know about tapering depressed patients receiving antidepressants in primary care. The patients who were recruited to the study were judged by semi structured research diagnostic interviews to be in remission, but their status at the time of prescription of antidepressants is only reconstructed retrospectively. Primary-care physicians typically do not make decisions to prescribe antidepressants on the basis of a formal semi structured interview with symptom counts.

This article provides no report of the treatment received by patients assigned to the maintenance antidepressant group and what quality and quality of care they received or even their adherence at the start of the study. Some studies suggest that a substantial proportion of primary care patients are already nonadherent six months after their initial prescription of antidepressants.

NICE guidelines recommend maintenance therapy for two years, but these guidelines are based on experiences in tertiary specialty psychiatric populations where the course of depression may be more severe and the risk of relapse greater. Arguably, a descriptive observational study would’ve been less expensive and more informative if it assessed adherence, contact with providers, and symptoms and side effects of primary care patients who were prescribed antidepressants a year earlier.

Results of this study would be much more informative if there had been better description of the care received in the control group or, better, a comparison/control group that provided equivalent amounts of attention, contact time, and support. It could well be that a specialized depression care manager with less training could have achieved equivalent results. Moreover, we simply don’t know if physicians advising their remitted patients to taper antidepressants would have achieved the same results.

On 2015 Apr 25, Martin Turner commented:

This should be an interesting study, but I suspect the results may be ambiguous. I think you may be giving your diet group too much sodium. 3 g of sodium is 130 mmol which is close to average sodium intake. Extracellular fluid expands when sodium intake exceeds approximately 800 mg or 35 mmol/day (Freis. Circulation 53(4):589-595 1976). This study will require a substantial effort and it would be worth reducing the sodium intake of your diet group to approximately 800 mg or 35 mmol/day.

On 2015 May 02, Donald Forsdyke commented:

DIABETIC PLASMA IS MORE ROULEAUGENIC THAN NORMAL PLASMA This fine paper reports an impressive in vivo method for evaluating rouleaux formation. The authors correctly state that rouleaux formation is "attributed to some changes in the plasma concentration" of certain proteins, "which modify the interaction between RBC." However, they go on to conclude that "diabetic erythrocytes have a higher propensity to form aggregates." To show this they would have had to study both RBCs from diabetic patients in normal subjects’ plasma and the converse (normal RBCs in diabetic patients’ plasma). But they do not report such experiments.

Following their original premise, it would be predicted that, when suspended in plasma from diabetic patients, normal RBC (of the same blood group) would form rouleaux just as well as the RBC from diabetic patients. That the primary change is in the surrounding plasma has long been known. Indeed, normal plasma can be made rouleaugenic by either heating to generate polymeric albumin, or merely by slightly concentrating. In both circumstances, the plasma will aggregate autologous RBC (1).

It appears that the aggregation is entropy-driven, showing a degree of specificity (like-RBC aggregating with like-RBC) analogous to the homoaggregation of macromolecules that can be induced by increasing the concentrations of surrounding but dissimilar macromolecules (2). The early history and theoretical implications of rouleaux formation are reviewed elsewhere (3), and in my webpages: see Entropy-Driven Protein Self-Aggregation at http://post.queensu.ca/~forsdyke/mhc001.htm

(1) Forsdyke DR, Palfree RGE, Takeda A (1982) Formation of erythrocyte rouleaux in preheated normal serum: roles of albumin polymers and lysophosphatidylcholine, Canadian Journal of Biochemistry 60: 705-711.

(2) Forsdyke DR, Ford PM (1983) Segregation into separate rouleaux of erythrocytes from different species. Evidence against the agglomerin hypothesis of rouleaux formation. Biochemical Journal 214: 257-260.

(3) Forsdyke DR (1995) Entropy-driven protein self-aggregation as the basis for self/not-self discrimination in the crowded cytosol. Journal of Biological Systems 3: 273-287.

On 2016 Oct 19, David Nunan commented:

We wrote an editorial on this open access editorial but ours was put behind a paywall. In addition, the open access editorial also received the following: - numerous promotional tweets from the @BJSM account (ours received one at the time of its release) - a podcast https://soundcloud.com/bmjpodcasts/you-cant-outrun-a-bad-diet-draseemmalhotra-on-weight-loss-strategies - Reference in a blog http://blogs.bmj.com/bjsm/2015/05/21/its-not-just-bjsm-talking-about-healthy-nutrition-real-food/

To attempt to redress the balance, here we provide an abriged pre-published version of our editorial piece for interested readers.

Malhotra and colleagues highlight the important health benefits of physical activity in cardiovascular disease, type 2 diabetes, dementia and cancer but dismiss any benefit for weight loss in obesity.1 However, the poor use of existing evidence to underpin this dismissal and the resultant press release generated headlines that send a misleading message to patients and the public.

Poor use of evidence.

An evidenced-based approach to clinical decision making involves understanding the best type of evidence to use, how to appraise it, and assessing its usefulness in practice. Such an approach determines the risk of bias and ensures decisions are guided by accurate, meaningful evidence. The GRADE guidelines cite expert opinion alone as very low quality evidence, with a high risk of bias.2 Unfortunately much of the evidence cited by Malhotra and colleagues was based on expert opinion or was not referenced. For example, when excluding a relationship between obesity and physical activity, the evidence cited was an opinion piece by Luke & Cooper3, which has been criticised for ignoring observational and experimental studies that support a relation between physical activity and obesity. They give no reference for their claim that according to the Lancet Global Burden of Disease report poor diet now generates more disease than physical inactivity, alcohol, and smoking combined.

Current recommendations for physical activity in obese patients.

To substantiate their dismissal Malhotra and colleagues would have to provide better evidence to justify why current guidance is wrong. The World Health Organisation currently recommends that obese and overweight individuals engage in regular physical activity (60 minutes a day for children and 150 minutes per week for adults).4 National Institute for Health and Care Excellence (UK) guidance advises health professionals should support obese patients in increasing their physical activity, particularly through activities that fit easily into their everyday life.5 Recognising the value of physical activity in promoting a healthy lifestyle, including in obese patients, the Academy of Medical Royal Colleges recently issued its report Exercise: the Miracle Cure, which summarised the evidence on physical activity and called on doctors “to promote the benefits of regular physical activity to their patients and to communities in their wider roles as ‘advocates for health’.”6 They also highlight systematic review evidence that body weight may stay the same despite a reduction in high risk visceral fat as a result of increased muscle mass.

Current evidence for physical activity and obesity.

The Malhotra editorial challenges the effectiveness of physical activity for weight loss in obese individuals. The guidelines and reports highlighted above present much of the evidence base and include a specific Cochrane systematic review by Shaw and colleagues7 which was ignored by the Malhotra editorial. This review included 43 studies reporting results from 41 randomised controlled trials including a total of 3476 participants. The duration of included studies ranged from 3 to 12 months (61% less than 4 months), including follow-up. Although methodological quality for included studies was variable according a risk of bias assessment, the Cochrane review authors report that exercise combined with diet resulted in greater weight reduction than diet alone (mean difference -1.0 kg; 95% confidence interval (CI), -1.3 to -0.7). They also note the finding that increasing exercise intensity increased the magnitude of weight loss (mean difference -1.5 kg; 95% CI -2.3 to -0.7). An additional finding was that exercise improved a range of secondary outcomes even when weight loss did not occur. The authors conclude exercise was an effective weight loss intervention, particularly when combined with dietary interventions but make no inference to the poor quality of included trials and little reference to the fact that diet alone appeared better than exercise alone (mean difference -3.6 kg, 95% CI -4.3 to -3). Whether these differences are meaningful requires good clinical judgement and open, honest discussion.

These outcome and quality findings are repeated in a more recent systematic review assessing treatment (including diet, exercise or diet and exercise) for overweight and obesity in adult populations.8 When objectively assessing the best available evidence, for which we acknowledge the majority is of low quality, it would appear that diet alone (mainly low fat) is better than exercise alone but diet and exercise both are better than either. A further key point is that improvements in important health risk factors are observed with exercise even without weight loss.

Conclusions

There is little doubt that diet is key in the management of the obesity epidemic. What is less helpful, and does no-one, including the public and patients, any favours, is the production of opinion pieces that use evidence poorly or not at all.

References 1. Malhotra A, Noakes T, Phinney S. Editorial: It is time to bust the myth of physical inactivity and obesity: you cannot outrun a bad diet. Br J Sports Med doi:10.1136/bjsports-2015-094911 2. Howard Balshem Mark Helfand, Holger J. Schünemann, Andrew D. Oxman, Regina Kunz, Jan Brozek, Gunn E. Vist, Yngve Falck-Ytter, Joerg Meerpohl, Susan Norris, Gordon H. Guyatt. GRADE guidelines: 3. Rating the quality of evidence July 30, 2010; Published Online: January 06, 2011 3. Luke A, Cooper RS. Physical activity does not influence obesity risk: time to clarify the public health message. Int J Epidemiol 2013; 42(6): 1831-6. 4. http://www.who.int/mediacentre/factsheets/fs311/en/. Accessed 28 April, 2015 5. http://www.nice.org.uk/guidance/cg43/resources/guidance-obesity-pdf. Accessed 28 April, 2015 6. http://www.aomrc.org.uk/general-news/exercise-the-miracle-cure.html. Accessed 28 April, 2015. 7. Shaw KA, Gennat HC, O’Rourke P, Del Mar C. Exercise for overweight or obesity. Cochrane Database of Systematic Reviews 2006, Issue 4. Art. No.: CD003817. DOI: 10.1002/14651858.CD003817.pub3. 8. Peirson L, Douketis J, Ciliska S, Fitzpatrick-Lewis D, Usman Ali M, Raina P. Treatment for overweight and obesity in adult populations: a systematic review and meta-analysis CMAJ Open. 2014 Oct-Dec; 2(4): E306–E317. Published online 2014 October 1. doi: 10.9778/cmajo.20140012.

On 2016 Jan 04, Amanda Capes-Davis commented:

Anyone who uses MDA-MB-435 to study breast cancer needs to be aware that MDA-MB-435 comes from the same donor as M14, a widely used melanoma cell line. Although the origins of MDA-MB-435 and M14 continue to be debated, a strong body of evidence exists to say that MDA-MB-435 is misidentified and is actually melanoma.

Concerns regarding the origin of MDA-MB-435 were first raised by Ross et al., 2000 (PMID 10700174). This cDNA microarray analysis of the NCI-60 cell line panel showed that MDA-MB-435 closely resembled melanoma cell lines rather than breast. Rae et al. (PMID 15679052, 17004106) later demonstrated that MDA-MB-435 shares a common origin with the M14 melanoma cell line. Numerous samples of both cell lines have been tested with the same finding.

This finding has been debated based on phenotypic evidence. The article above notes that MDA-MB-435s expresses LHRHR. MDA-MB-435 has previously been shown to express other breast-specific markers e.g. Sellappan et al., 2005 (PMID 15150101). However, it should also be noted that MDA-MB-435 expresses melanoma-specific markers e.g. Ellison et al., 2002 (PMID 12354931).

Phenotype can be helpful but is often confusing when looking at misidentified cell lines. For example, cell lines that are cross-contaminated with HeLa have been shown to express liver- and other tissue-specific markers, leading to the incorrect assumption that cell lines are authentic when they are actually HeLa derivatives e.g. Chang, 1978 and response by Nelson-Rees, 1978 (PMID 622561). It is important to make any phenotype-based decisions using a broad expression profile and not rely on one or two individual markers.

It is also essential to test for donor origin using a method such as short tandem repeat (STR) profiling. STR profiling makes it clear that MDA-MB-435 and M14 share a common donor origin; review of their origins show that one of these cell lines must be misidentified.

Donor origin continues to be a subject of debate for MDA-MB-435 e.g. Chambers, 2009 (PMID 19549886). Users of MDA-MB-435 must make decisions based on available data. In my view, MDA-MB-435 is an unsuitable model to study breast cancer and the model described above is more likely to be suitable for melanoma research.

On 2017 Nov 24, Amanda Capes-Davis commented:

Following up from the comment by Deilson Elgui de Oliviera, ICLAC has new data on MDA-MB-435 and its origin (https://www.ncbi.nlm.nih.gov/pubmed/28940260).

MDA-MB-435 came from the same donor as M14, a cell line that was established from a male with melanoma. Most publications have concluded that MDA-MB-435 is misidentified but the topic has been a controversial one. MDA-MB-435 can express breast-specific markers and its karyotype is XX, which would fit with its reported origin from a female with breast cancer.

ICLAC went back to the original publications for these two cell lines and found that M14 was established and submitted for publication before MDA-MB-435. The originator, Dr Donald Morton, established a specimen repository at the John Wayne Cancer Institute (JWCI); we contacted JWCI to ask if early specimens were available. JWCI was able to supply early samples of the M14 cell line, a lymphoblastoid cell line from the same donor (ML14), and serum from the melanoma donor dating back to 1973. The originator's work made subsequent testing possible.

Testing of these early samples showed that MDA-MB-435 came from the M14 donor, proving that MDA-MB-435 is misidentified. The XX karyotype was caused by chromosomal rearrangement; the donor's lymphoblastoid cell line is XY.

This "Tale of Two Cell Lines" is a cautionary tale showing that phenotype can be misleading when working with cell lines. Always test your cell lines for authenticity. A cell line is itself a variable that can never be taken for granted.

On 2017 Oct 24, Deilson Elgui de Oliveira commented:

MDA-MB-435 devrivatives are listed as contaminated cell lines (M14 Human melanoma, CVCL_1395) in the ICLAC Database of Cross-Contaminated or Misidentified Cell Lines (Version 8.0, Publication Date 1/12/2016).

On 2016 Aug 23, Ben Goldacre commented:

This trial has the wrong trial registry ID associated with it on PubMed: both in the XML on PubMed, and in the originating journal article. The ID given is NCT0131821. We believe the correct ID, which we have found by hand searching, is NCT01318213.

This comment is being posted as part of the OpenTrials.net project^[1] , an open database threading together all publicly accessible documents and data on each trial, globally. In the course of creating the database, and matching documents and data sources about trials from different locations, we have identified various anomalies in datasets such as PubMed, and in published papers. Alongside documenting the prevalence of problems, we are also attempting to correct these errors and anomalies wherever possible, by feeding back to the originators. We have corrected this data in the OpenTrials.net database; we hope that this trial’s text and metadata can also be corrected at source, in PubMed and in the accompanying paper.

Many thanks,

Jessica Fleminger, Ben Goldacre*

[1] Goldacre, B., Gray, J., 2016. OpenTrials: towards a collaborative open database of all available information on all clinical trials. Trials 17. doi:10.1186/s13063-016-1290-8 PMID: 27056367

* Dr Ben Goldacre BA MA MSc MBBS MRCPsych<br> Senior Clinical Research Fellow<br> ben.goldacre@phc.ox.ac.uk<br> www.ebmDataLab.net<br> Centre for Evidence Based Medicine<br> Department of Primary Care Health Sciences<br> University of Oxford<br> Radcliffe Observatory Quarter<br> Woodstock Road<br> Oxford OX2 6GG

On 2017 Dec 18, Shaun Khoo commented:

Green Open Access: The accepted manuscript is available from the UNSW Institutional Repository.

Open Data: The underlying dataset for this paper is available on Figshare.

On 2016 Feb 23, Jomar Rabajante commented:

Erratum (typographical error in the main text during publisher formatting): j=dj should be dj=d

In fact, the equation in the Supplementary Information is correct. The error does not affect or change the information in the figures, the results and the discussion in the manuscript.

On 2015 Apr 26, Donald Forsdyke commented:

The data in this interesting paper seem not incompatible with the hypothesis that the need to prevent recombination with other organisms drives a organisms GC%. This anti-recombination selective effect (resulting in the reproductive isolation needed for maintaining species integrity) is something the entire organism has to adapt to. Having adapted, it seems not unlikely that, in some cases, an artificial changing of GC% (as in the Kelkar paper) would be deleterious. This would be particularly evident in the case of 'lower' species that had not superimposed other mechanisms for maintaining reproductive isolation. Absence of superimposed mechanisms would prevent GC% values from seeking new equilibrium positions. For more see my text Evolutionary Bioinformatics (Springer, New York, 2011).

On 2016 Jun 23, Francisco Felix commented:

Unfortunately, methodology section seems to be faulty in this publication. Some of the information I did not find when reading it: 1. What was the administration scheme of naringin (once daily, weekly, other)? 2. What was the dose of cisplatin used (as unlikely as it seems to be, the authors really have forgotten to report it)? 3. Have the behavioral tests been done 48 or 24h prior to animal killing? Both contradictory information were given. I am afraid this paper is not really useful without clarification upon these issues, once they are critical for the methodology the researchers used. Maybe editorial and peer reviewing process could have been a little more rigorous in this case? I disclose no conflicts of interest.

On 2015 Apr 25, David Keller commented:

Unfortunately, this study did not measure any relevant clinical outcomes, such as weight loss

This study reports that subjects who consumed thlylakoids (a spinach extract) had reduced hunger, increased satiety, reduced cravings for snacks, and other purely subjective measures. There was no reported improvement in clinically important outcomes, such as: actual weight loss; body fat composition; waist-to-hip ratio; lipid profile; blood sugar levels, and so on. It is common knowledge that consuming French pastries also reduces hunger, increases satiety and reduces cravings for snacks. What is needed is an intervention proven to reduce subjective cravings for food while promoting actual weight loss and other outcomes which directly affect health. This study does not present any clinical evidence to support the authors' recommendation that thylakoids be used as a food supplement.

On 2017 Mar 07, Atanas G. Atanasov commented:

Nicely summarized information, which is of interest both for researchers and for consumers. I have highlighted this review at: http://healthandscienceportal.blogspot.com/2017/03/functional-components-and-medicinal.html

On 2015 Aug 31, Respirology and Sleep Journal Club commented:

This article was reviewed by the University of Toronto, Division of Respirology journal club as well as online using twitter. The author and others worldwide joined in for the discussion.

On 2015 Sep 20, Eythor Kristjansson commented:

The paper (Meisingset et al. BMC Musculoskeletal Disorders (2015) 16:56 DOI 10.1186/s12891-015-0517-2) reports the findings of a cross-sectional study, which examined set of tests to investigate cervical motor control. One of the tests that were used was the Fly Test©, which Meisingset et al cited “as described by Kristjansson et al., 2010” [1]. When I read the abovementioned article by Meisingset et al., [1] I found out that their article does not replicate the study by Kristjansson & Oddsdottir 2010, at all [2]. There are two main crucial differences in Meisingset et al. study and the original study by Kristjansson & Oddsdottir, 2010 [2], which make these 2 studies incomparable. First, there are several inbuilt parameters in the patterns in the original Fly Test©, and it is impossible for Meisingset et al., to know what inbuilt parameters were used and how we used these parameters. Meisingset et al. just copied the geometry of 2 patterns without permission from the publisher. Second, Meisingset et al. methodology is completely different from ours. Research results using different inbuilt variables within the patterns and completely different methodology can never be comparable. It is therefore not surprising that Meisingset et al. study found no difference in the performances between controls and neck pain group, besides that the controls performed worse in one sequence of what Meisingset et al., misleadingly call “the Fly Test”. Following are the main differences in methodology of the 2 studies: 1. Set-up: Our original study used a seated position, 1m from the screen whereas Meisingset et al. used standing position, 2.5m from the screen. It can be easily proven mathematically that different angles are produced on the screen depending on the patients distance from the screen. This alone will generate different results as longer distance (2.5 m) will produce narrower angle on the screen i.e. lesser ROM than a distance of 1m, as we used in our original paper. 2. Number of patterns and repetitions: Our original study used three patterns, easy-medium-difficult of uniform size and three repetitions for each pattern in random order, altogether 9 trials. Meisingset et al. used two patterns, easy and medium with 2 trajectory sizes; on smaller and one bigger. Furthermore, the control group in Meisingset et al study just performed the patterns once, while the neck pain group performed the patterns twice, not in random order. 3. Duration of patterns: Our original study used 25, 40, and 50 seconds for the easy-medium-difficult patterns, respectively whereas Meisingset et al. used 30 seconds for both patterns; easy and medium. Here, Meisingset et al. wrongly cited our original paper; see Table 1 in their study [1]. The velocity in Meisingset et al. study for the medium pattern must therefore be faster than in the easy pattern as longer distance is traveled in the medium pattern. I have one more serious concern about Mesingset et al. conclusion regarding what they misleadingly call “the Fly Test” in their study: In the Discussion section, Meisingset et al. cited Oddsdottir et al. 2013, regarding gender difference. Meisingset et al. write on p. 11 under the heading Trajectory movement control: “In the HC group (Healthy Control) we found that females had a consistently larger deviation in the Fly test compared to men in the HC group suggesting that the group differences in Kristjansson & Oddsdottir study, 2010 may be influenced by the different gender distribution in the groups”. Meisingset et al. continue: “However, Oddsdottir et al. found similar results between healthy females and men for the same Fly test [3]”. “This discrepancy between the studies requires further investigation before this test can be implemented in clinical practice.” Meisingset et al. are making reference to our database study [3]. The result section on p. 3 in our database study starts on the following statement: “The multivariate analysis of variance indicated a statistically significant effect for age (p < 0.001), but not for gender (p > 0.05). Therefore normative data were only distributed according to age”. It is impossible for us to understand how Meisingset et al. can cite our work in such a wrong way. References [1] Meisingset I, Woodhouse A, Stensdottir A-K, Stavdahl Ö, Lorås H, Gismervik S, Andersen H, Austreim K, Vasseljen O, Evidence for a general stiffening motor control pattern in neck pain: a cross sectional study. BMC Musculoskeletal Disorders (2015) 16:56 DOI 10.1186/s12891-015-0517-2 [2] Kristjansson E, Oddsdottir G."The Fly" - A new clinical assessment and treatment method for deficits of movement control in the cervical spine: reliability and validity. Spine 2010; 35: E1298–E1305 [3] Oddsdottir G, Kristjansson E, Gislason M. Database of movement control of the cervical spine: Reference normal of 182 asymptomatic people. Manual Therapy 2013; 18: 206–210.

Competing interests: EK is the founder of NeckCare Ltd, a start-up innovation company developing the Fly Method©, and other innovative methods, for diagnostic and treatment purposes. The tasks created in the new Fly Method© are protected by patent rights.

Meisingset I, 2015 http://www.nih.gov NIH Home

On 2016 Jan 12, Elaine Beller commented:

This is a topic that has been debated on my Facebook feed recently. This group has performed a RCT, but reports within-group changes throughout, and not the between-group results. Some of the within-group results have P-values less than 0.05, and these are the ones that been reported in the abstract. This approach implies that there were no changes in the control group, and that both groups had similar baseline values (which is not so). I suggest the authors should give the results for the comparison between groups on final value, adjusted for baseline (ANCOVA).

Bland and Altman have written several excellent artcles explaining why the within-groups approach should not be used. In one which is open-access, they say: "This approach is biased and invalid, producing conclusions which are, potentially, highly misleading. The actual alpha level of this procedure can be as high as 0.50 for two groups...",

Comparisons against baseline within randomised groups are often used and can be highly misleading. Trials 2011, 12:264 doi:10.1186/1745-6215-12-264

On 2016 Dec 15, Morten Oksvold commented:

This article has been retracted due to several problematic issues with data duplications.

This case represents an excellent example for how an editorial and review process should NOT be performed:

1. Apparent extensive duplication of data was not detected during the review process.
2. Three issues of apparent duplications were reported to Science by Johannes M Dijkstra June 25th 2015.
3. Dijkstra reported to Science regarding a fourth apparent data duplication August 13th.
4. Science confirmed that they will publish a correction of the reported duplications September 24th.
5. Science published a correction October 23rd 2015.
6. The published correction poorly describes the details of the correction and the general explanation for how the published duplications were made is difficult to understand.<br>
7. The extensive correction was not discussed with the authors' research institutes.
8. Apparent data duplication in the corrected version was reported November 18th.
9. Science published an expression of concern December 10th. Imperial College London investigates the case.
10. Science finally retracted the article one year later (December 10th).
11. The published retraction is only accessible for subscribers.

Please see the full discussion at PubPeer:

https://pubpeer.com/publications/B74CF2D21C4A180A5685A30DC06D29#fb113276

On 2015 Sep 04, Bill Cayley commented:

A good example of when LESS may be MORE for treating radiculopathy: https://lessismoreebm.wordpress.com/?s=radicular

On 2015 Jul 17, Christopher Southan commented:

The authors should please make the specific molecular structures for the inhibitors in table 2 available (SMILES, InChI or PubMed CIDs). A figshare sheet would be an option.

On 2016 May 27, ALAN HOWE commented:

There appears to have been an 'autocorrect'-style error with this citation & paper, as 'c-MET' (aka the hepatocyte growth factor receptor) was erroneously changed into 'c-mesenchymal-epithelial transition' in several places.

On 2016 Mar 23, Laila Poisson commented:

Egon, you are correct, we used the Human Metabolome Database (HMDB) number to map identified metabolites into pathways. Thank you for pointing out this typo.

On 2015 Dec 25, Egon Willighagen commented:

Can the authors confirm that they used the Human Metabolome Database, or provide a website or reference for the "Human Metabolon Database" cited in the paper, please?

On 2015 Jun 26, Torsten Seemann commented:

The latest version of this software is available as a Github repository at https://github.com/aakrosh/indelMINER

On 2016 Mar 03, Farrel Buchinsky commented:

An interesting article about 21 children noting that there was a correlation between the number of surgeries and the preceding delay between symptoms developing and a diagnosis being made. This finding would be compatible with the possibility that the longer one delays the more a child is likely to need more surgeries. If that is true then that should encourage us to make the diagnosis as soon as possible so that we can operate as soon as possible and so that the child could undergo fewer surgeries rather than more surgeries. Given that this was not a randomized trial would the data also be compatible with another possibility? Could it be that children with the least delay may have gone into spontaneous remission had their delay continued and they had undergone no surgery?

Two other observations struck me when I looked at the data. There was a statistically significant association between the age of symptom onset and the time between symptom onset and diagnosis. Children whose symptoms began before they were 5.5 y (66 months) had a median delay of 33 months and the children who were older than 5.5 y when their symptoms began had a median delay of 3.5 months. Many case series have demonstrated an association between younger age being associated with a more aggressive clinical course presumably resulting in greater number of surgeries. That was not seen in this group of 21 patients.

Can you please check the statement, "The average time from symptom onset to diagnosis was 52.3 months". I calculate it to be 20.3 months. I used the following values: 1 3 24 7 33 36 48 3 4 4 54 7 21 26 38 3 43 59 1 1 10

On 2015 Jun 18, Elsamma Chacko commented:

This talks about how very intense exercise can increase glucose levels. Light to moderate exercise is better for lowering glucose peaks.