On 2019-05-22 23:31:32, user Charles Warden wrote:

I thought this part was interesting:

"How much does it actually cost to run a workshop like EDAMAME? The first year, we ran the workshop for less than $14,000; students paid their own expenses of room and board; and no workshop fees were charged."

On 2019-05-22 20:23:29, user Salvatore Pappalardo wrote:

The manuscript was finally published on PLOS ONE, on May 1st 2019. <br /> It is possible to access and download the full paper: https://journals.plos.org/p...<br /> Salvatore E Pappalardo, corresponding author.

On 2019-05-22 14:18:52, user A. S. wrote:

When you write "Mitochondria evolved from archaebacteria", you are probably confusing alphaproteobacteria (which are Eubacteria) with archaebacteria.

On 2019-05-22 12:14:15, user John Novembre wrote:

This looks very interesting and I hate to be the person to complain about overlooked work but... I’d love to see comparisons to prior work on an effective number of codons to deal with varying GC content (Nc’, from “Accounting for Background Nucleotide Composition When Measuring Codon Usage Bias” MBE 2002). I think your approaches likely have advantages, but a comparison, or at least a few more words of discussion, would be helpful. Happy to chat via email too if it’s helpful. John Novembre

On 2019-05-21 21:39:25, user Charles Warden wrote:

I see that you mention "precisionFDA" in the discussion, but I think having continual comparisons overtime (from individual companies or labs) has benefits.

For example, these are my experiences (with Veritas WGS data and Genos Exome data):

http://cdwscience.blogspot....

Also, while you can specify the regions to compare (like I did for CDS regions), there are also other apps where you can get more details reports.

For example, Vcfeval + Hap.py Comparison was suggested to me, and it helped me troubleshoot error messages to get the main variant comparison function working.

On 2019-05-21 19:32:06, user Emma wrote:

I'm not allowed to post the accepted manuscript here, but there were some important corrections. There's a copy at: http://eeg.github.io/lab/_s...

On 2019-05-21 19:05:51, user Michele Busby wrote:

Nice paper! One thing I think I didn't do very well in my application was how I handled library complexity. I think I modeled it as if you could sequence forever and keep getting more reads. Did you look at how library size affects the saturation?

On 2019-05-19 23:04:36, user Charles Warden wrote:

Thank you for putting together this pre-print.

I am sure that there are some situations where higher read coverage can be beneficial. Admittedly, I think other applications like mutation calling would have a relatively greater need for more reads (and that would depend upon the evenness of coverage for your library type, and possibly what genes you have the greatest need to check mutations in), but I think it is perfectly reasonable to focus on the differential expression part for one paper.

That said, when I saw the tweet mentioning "We find > 70% published studies would have benefitted from increasing number of reads sequenced", I was a little worried about the influence it could have on readers for the following reasons:

1) If somebody is considering purchasing a Desktop sequencer for RNA-Seq analysis, I think 2-6 Million reads to cover genes with above average expression may be a better option than using targeted gene panels. For example, if you do re-analysis (with unique read counts and updated differential expression methods, like DESeq2, limma-voom, etc.), I think the MiSeq data from the cuffdiff2 paper shows reasonable results (for treatments with clear gene expression changes).

2) In most cases, I am more concerned about people having replicates than needing more reads (at least for gene expression).

I apologize that I think it may be a little while before I can focus more on point #1, but I tried to take a quick look at this paper.

I think that it is great that you performed benchmarks with DESeq2, edgeR, and limma-voom (although maybe you want to change “limma” to “limma-voom” in the abstract?). I apologize for not being able to find this on the superSeq page (although I did find the reminder for the previous biocLite() command for dependencies to be helpful), but are tables of pre-processed counts (and their gene lists with all 3 methods) readily available for the 1,021 contrasts?

I am also glad that you are looking at differentially expressed gene counts (and not just unique read sequences) for your rarefaction plot, since I think that is a more relevant measure for whether you get functionally relevant results. However, in terms of the vignette example, I think the difference between 1338.968 “Estimated number of discoveries” at read depth of 1 and 1888.286 “Estimated number of discoveries” at read depth of 3x is within the range that could be achieved from changing the p-value method and/or changing the FDR cutoff (from 0.05 to 0.25 or 0.50, for example).

Similarly, I am concerned about some of the maximum gene counts in the pre-print, which look like pretty much the entire genome in Figures 2 (and are already above 2000-4000 genes in the theoretical example in Figure 1). I think the best balance for functional enrichment is often around 1000-2000 total genes (~5-10% of genes). So, I would be interesting in knowing if your framework can answer a question like “What is the range of reads needed to identify 1000 or 2000 differentially expressed genes?.”

While some treatments have greater effects, I think 10-20 million reads for a human polyA library is probably usually OK (and perhaps double that for a ribosome-depleted library, with a lower exonic percentage). I think that is pretty much what Figure 1A shows (although that looks like close to 30 million reads), but I am wondering if there is a figure derived from your ~1000 comparisons (and/or a parameter that can be added to plot pre-computed values in the R package).

Also, am I correctly understanding that you downloaded pre-processed counts? Did you look at some of the most extreme differences and test reprocessing the samples to see if that helped the differentially expressed gene counts become more similar? There are situations where I would prefer to start from FASTQ files and process all samples the same way.

For example, 60,000 in Figure 5 seems like it probably includes transcripts – is it possible to only look at unique gene-level counts (that is admittedly what I would be interested in checking)? Or, are there outliers that can be excluded if you only look at human and mouse experiments (trying to control for annotation effects)? Also, I’m not hugely concerned about the annotation in model organisms like yeast or fly, but the total number of genes in the genome is going to have some effect (both in terms of the effect on the differential expression models, as well as having very different genome sizes and maximum gene counts).

Finally, going back to my original point #2, I would expect replicates should help reduce false positives. With large enough sample sizes, I would expect to pick up more subtle effects. However, with 1-3 replicates, I think fewer genes to narrow down candidates may be beneficial (rather than increasing the number of genes identified). For example, at an estimated FDR of 0.05, how many genes are identified between biological replicates for the same group (to see if increased sensitivity may actually be affecting the accuracy of the estimation to allow more false positives, which seems likely if you are identifying >20% of the genome, in my opinion).

Or, it is a slightly different point, but I think 6 replicates are used in Figure 3A. If 6 replicates exist for an experiment, what is the effect of having 3 replicates at the current coverage versus 6 replicates at halved coverage? Sometimes, getting people to even do comparisons with triplicates can be a challenge.

I apologize that this is kind of a long comment, but that is because I think this is an important topic. When I get the point of being able to post some pre-prints, I realize that answering questions from long commenters can take time, but I think that is very important for the scientific community (in terms of helping put together the best possible paper for peer-review).

On 2019-05-21 12:17:17, user Brendan Barrett wrote:

now in press: https://www.nature.com/arti...

On 2019-05-21 10:33:38, user Annkatrin Bressin wrote:

Hi, I'm really interested in the data. Is the data publicly available? Thank you in advance.

On 2019-05-20 18:59:04, user Aaron Gitler wrote:

Comments on Bolognesi et al.

By Aaron Gitler and Steven Boeynaems (Stanford University)

Many human neurodegenerative diseases are associated with protein aggregation. In Alzheimer disease, Amyloid-beta (A) and tau form aggregates, in Parkinson disease, synuclein forms aggregates, and in Huntington disease, a form of huntingtin with an expanded polyglutamine tract forms aggregates. Whether or not these aggregates cause disease is an area of intense study, with important therapeutic implications. For example, if the aggregates are what is causing disease, then strategies to dissolve or degrade the aggregates would be warranted. But if the aggregates were potentially protective, then therapeutic strategies to promote aggregate formation of toxic entities would be warranted. For the neurodegenerative diseases ALS and FTD, the RNA-binding protein TDP-43 is mislocalized to the cytoplasm and forms aggregates in the degenerating neurons of patients with these diseases.

TDP-43 harbors a prion-like domain, which is able to drive liquid-liquid phase separation. The liquid phases formed by TDP-43 are thought to lead to the formation of solid insoluble aggregates. Several mutations in the gene encoding TDP-43 have been identified in familial and sporadic ALS patients. Almost all of these mutations are located in the prion-like domain. Some reports have indicated that these pathogenic mutations can increase the aggregation-propensity of TDP-43. Whether the loss of TDP-43 from the nucleus (where it regulates a diverse array of mRNA targets) or rather its accumulation in the cytoplasm in an aggregated form (perhaps causing dysregulation of other RNAs and RNA-binding proteins) causes ALS (i.e., loss of function vs. gain of function) is still unresolved and an area of intense interest.

In this manuscript by Bolognesi and colleagues, the authors performed a systematic analysis of the prion-like domain of TDP-43 – engineering over 50,000 mutations – assessing effects on aggregation and toxicity in a yeast model. They made several key findings:

1) Mutations that change hydrophobicity and aggregation are predictive of changes in toxicity<br /> 2) Surprisingly, the mutations they introduced that increased hydrophobicity and aggregation actually decreased toxicity! <br /> 3) The mutations that increase the liquid-like properties of TDP-43 increased toxicity<br /> 4) Using a clever analysis of double mutant combinations they were able to infer the existence of specific structures within TDP-43’s prion-like domain.

Overall, this is a very exciting new story that provides a comprehensive and deep analysis of TDP-43 prion-like domain with surprising findings that mutations that increase aggregation actually decrease toxicity and the ones that increase the liquid phases of TDP-43 are the more toxic ones. This study has important implications for thinking about how TDP-43 contributes to ALS and FTD and also provides a framework and new method to study other neurodegenerative disease-associated aggregating proteins.

We have several comments and suggestions for the authors to consider.

1) RNA-binding seems to be required for TDP-43 toxicity in yeast (PMID: 18434538 and PMID: 20740007). The authors focused their mutational scan on the prion-like domain of TDP-43, which does not include the two upstream RNA-recognition motifs (RRMs). The specific mutations they introduced to the prion-like domain could potentially influence the ability of TDP-43 to bind RNA. It would be interesting for the authors to test several of their mutations for impact on RNA-binding. The binding site for TDP-43 (UGUGUGUG) has been characterized and several of its targets are known. It is possible that the mutations that decrease toxicity do so by decreasing TDP-43’s ability to bind RNA. This result would still be interesting and important and would provide insight into potential communication between the prion-like domain and the RRMs. The authors could also introduce additional mutations in the RRMs (e.g., FW) in the context of their most toxic mutants in order to define if the enhanced toxicity of these new mutations depends on RNA-binding or not.

2) Kaganovich, Kopito, and Frydman have described distinct subcellular quality control compartments, to which distinct misfolded proteins are routed (PMID: 18756251). These include a juxtanuclear compartment (JUNQ) and a peripheral perivacuolar compartment named insoluble protein deposit (IPOD). Proteins in the JUNQ are more mobile than those in the IPOD. Toxicity of a model misfolded protein can be reduced by directing it away the JUNQ and towards the IPOD (PMID: 22967507). The authors should consider analyzing whether their toxic and non-toxic mutants accumulate in yeast JUNQs or IPODs, using the established markers for these compartments.

3) Expression level of TDP-43 strongly correlates with toxicity in yeast (and other systems). Surprisingly, the authors see (in Extended Fig. 5b), that the toxic mutants are expressed at much lower levels than the non-toxic ones. This result is unexpected and it would be helpful to extend the analysis to additional mutants, and potentially longer time of protein induction (e.g., >8 hours).

4) While recurrent ALS mutations in TDP-43 were in general moderately more toxic when expressed in yeast, they do not necessarily associate with increased charge or decreased hydrophobicity. Rather, they seem enriched for G/A/N to S/T mutations (5/12). While such mutations could increase hydrophilicity substantially via phosphorylation, these human disease mutations do not replicate the pattern observed for the very toxic yeast mutations. An explanation why “severe” mutants are not observed in humans is that such mutations are potentially not tolerated during development, leaving only the possibility for moderate mutations to cause age-related disease. Hence, while the signature of increased hydrophilicity/decreased hydrophobicity is clearly associated with toxicity in yeast, this finding does not necessarily illuminate the biophysics behind TDP-43 toxicity in humans.

5) Some of the well-established TDP-32 ALS mutants have been previously studied in vitro, where authors have shown that these mutants perturb liquid phase separation of TDP-43 (PMID: 27545621). It would be great if the authors could specifically image these mutants in their yeast system. Additionally, evaluation of key mutants, both ALS-relevant ones and the ones with strong yeast phenotypes, in a human cellular context would be crucial for interpreting these results. It is known that condensates may present with different material properties between yeast and human cells (PMID: 26238190), and yeast cells potentially lack the (non-)canonical chaperone systems that may alter TDP-43 phase behavior (e.g. PMID: 29677512, PMID: 30100264).

6) Currently, the round, dynamic, perinuclear TDP-43 condensates are described as being liquid(-like). However, this claim is only weakly supported. Micronscale dynamic liquid behavior should be observed before such claims can be made (i.e., plastic deformation, wetting, fusion/fission). As mentioned above, investigating the relationship of these condensates to IPOD/JUNQ will be important.

7) This paper once more provides evidence for the fact that aggregates may be beneficial by sequestering toxic protein species (e.g. oligomers, liquid droplets). Interestingly, a recent study identified stress-induced liquid-like wildtype TDP-43 condensates that were toxic to human cells (PMID: 30853299). However, it remains unclear what the effect of ALS-associated mutations is on the latter ones, and how they relate to the TDP-43 pathology in human patients. While these two studies suggest that indeed “liquid droplet toxicity” can indeed happen in yeast and human cell culture, it remains to be tested whether this relates to the disease. Especially, in the context of yeast: Overexpression of TDP-43 out of its endogenous context induces spontaneous cytoplasmic condensation and toxicity, which is generally not observed in human systems. This is likely caused by TDP-43 perturbing endogenous yeast RNA metabolism and proteostasis pathways. Any molecular perturbation (i.e. mutations) that make TDP-43 inert for yeast clients (i.e. aggregation in this study, RNA-binding mutants in PMID: 18434538 and PMID: 20740007), are expected to reduce toxicity. While loss of RNA binding is indeed beneficial in yeast, in the human system RNA binding protects against toxicity (PMID: 30826182). The latter observation suggests that toxicity in yeast –and the associated biophysical features of TDP-43– may not always be correlated with toxicity in human system

On 2019-05-20 08:22:53, user Jiri Hulcr wrote:

Hello Kirk et al.<br /> I am glad to see that you all are still interested in ambrosia beetles. With respect to the hypothesis about cycloheximide tolerance in some Ophiostomatales as related to their symbiosis with ambrosia beetles, I think it is important to consider that the tolerance is widespread in that fungal family, pre-dating their engagement with the beetles. I don't know the literature that well, but I would guess that cycloheximide tolerance is probably present in many Ophios that are not relevant to the beetles, or even compete with them. <br /> Have you measured whether the compound occurs in the galleries in quantities comparable to those in your media? That might get us close to a test whether it is related to higher fitness of the fungus, which would support your hypothesis. <br /> Thank you!<br /> Jiri Hulcr

On 2019-05-16 21:13:18, user Anasto wrote:

So much wishful thinking in one paper. Lots of claims of ‘proof’ where no such thing was shown. Here’s a list from reading just the first half: <br /> - 16S for identification of an actinobacteria strain is like from the 1990s. We know more about ribosomal sequence diversity these days. <br /> - Sampling effort is weak. The insects were pooled? - likely mixing environmental microbes. Not replicated over time and space. Cannot make claims about symbiosis from bugs caught in two traps. <br /> - The chitin media seem to be super low nutrient. The strep was isolated because this media was used, not because the bacteria is biologically meaningful. Why did you not use any other method? <br /> - The sampling from the nests was not really replication. A contaminant would turn up in exactly the same pattern if sampled from insects and nest material from the same three tubes. <br /> - The supposedly “mutualistic” streptomyces was detected in trace amounts (a few CFUs form large chunks of material or pooled bug samples). Why believe that it is biologically relevant? <br /> - The putative closest relative strep on genbank is a common soil bacteria. Makes me think that this really is a contaminant. <br /> - No attempt to work with the fungus of one of the insects. It was not even isolated or confirmed. Any conclusion about that system is moot. <br /> - Using some artificial agar and sawdust goo for the nests. Bound to be totally different than dead wood, and support/select an unnatural set of microbes.<br /> - I pull this one out verbatim first because it sums it all… line 381: “The consistent isolation of this Nectria sp. suggests that it is vectored by the ambrosia beetles”. There is a hundred years of literature showing that you cannot infer a vector, not to mention a symbiont, from just occurrences in media. You cannot ignore a century of research and throw around these vague claims. Nectria is common in sick plants, and there is no evidence that it interacts with these bugs, other than simply living in the same substrate. What if the bacteria and the Nectria are simply in the same habitat as the bugs, and THAT’s why they are in the samples? You need environmental control samples.

I stopped here – too many problems.

On 2019-05-19 13:54:30, user Ana Pedro wrote:

All the original data mentioned in this pre-print can be found by consulting the original data from Tucker and Pedro, F1000 Research (2018)

On 2019-05-19 03:06:39, user Mingye Wang wrote:

Now http://doi.org/10.1093/infd...

On 2019-05-18 23:32:15, user Juan David Leongómez wrote:

All comments regarding the ms and/or supplementary material (https://doi.org/10.17605/OSF.IO/9WKMT) are most welcome!

On 2019-05-18 23:25:12, user Mary Witt wrote:

Dear Authors,

Could you explain if some relationship could be observed between this manuscript and the following reported animal model?<br /> "Generation of a novel, multi-stage, progressive, and transplantable model of plasma cell neoplasms"<br /> https://www.nature.com/arti...<br /> Kind regards,

Mary Witt

On 2019-05-17 22:43:24, user Douglas Porter wrote:

This is the first author. Let me know if there is anything unclear in <br /> the protocol or there is any trouble with it. The method should really <br /> work as described out of the box pretty easily. I hope this is useful, <br /> especially to someone just starting out. There are many very excellent <br /> other CLIP protocols out there (eCLIP, iCLIP, PAR-CLIP, irCLIP), and this work is meant to provide its own particular usages (absolute <br /> quantification, ease of use, distinguishing RBPs, ect...), and insights.<br /> I hope it is helpful and you can contact me with any questions.

On 2019-05-17 19:29:22, user Sozanne Solmaz wrote:

Thanks Simon! Honestly, I am convinced that your model from the 2013 Liu et al. paper will prove to be 100% accurate.

On 2019-05-17 09:40:35, user Wouter De Coster wrote:

Dear authors,

Thank you for your interesting article. I would like to point you to an article which is worth discussing in this context: https://www.sciencedirect.c... <br /> Admittedly, I don't know much about machine learning but I am very enthusiastic about its application in large datasets in the context of genetics. However it seems that a better AUC is obtained using a more straightforward regression approach.

Regards,<br /> Wouter De Coster

On 2019-05-16 21:51:48, user Alberto Gonzalez wrote:

Italy1 influence needs to be reviewed, specially after the last paper from Olalde suggesting that modern Spanish people are in between Basques and Southern Italians since the Roman period.

On 2019-05-16 19:36:47, user Anuprita wrote:

Hi,<br /> I'm a Master's student and for my final project I'm interested in this dataset.<br /> Can you please tell me how can I get access to the Supplementary table information. I'm specifically interested in the Supplementary Table 6 information. In the supplementary information provided I was unable to find the full information in the supplementary tables. Please help.<br /> Thank you in advance <br /> Regards<br /> Anuprita

On 2019-05-16 19:18:39, user Rafael Rabelo wrote:

Article published at Acta Oecologica https://doi.org/10.1016/j.a...

On 2019-05-16 12:20:18, user Levi Yant wrote:

Note that correspondence is shared between me and Roswitha Schmickl.

On 2019-05-16 09:12:52, user Tim Vaughan wrote:

This article has been published: https://doi.org/10.1093/mol... (open access)

On 2019-05-16 03:06:12, user Craig Kaplan wrote:

Please note this paper entirely reevaluates results of Chen et al (2014). It may be helpful to shape your discussion about why your results are discrepant with interpretations of that paper. https://elifesciences.org/a...

On 2019-05-15 21:09:37, user Adam Taranto wrote:

This paper has got me thinking about what it means for a MITE to be "expressed" and if that is the same thing as not being suppressed/controlled.

Unlike retrotransposons, TIR family DNA-transposons (and their non-autonomous derivatives - MITEs) do not have a transpositional RNA intermediate. Though active transposition of MITEs does require the expression of a cognate TPase from an intact parent element elsewhere in the genome, it does not follow that all RNA-seq reads mapping to MITEs are derived from such elements.

I think there are plausible scenarios in which MITE sequences are transcribed but not themselves active.

1) Transcripts mapping to MITEs are derived from TPase-transcripts in parent elements.

Most TIR-family elements have internal Pol-II promoters for expression of their TPases. Transcripts from autonomous elements should map at least partially to an internal segment of a related MITE.

This option and probably be ruled out as no autonomous forms of MITE-Undine were found in Zt. Further, it looks like "TEtranscripts" requires a TE to be fully contained within a transcript to be counted. In this case parent-element derived transcripts might be excluded anyway.

2) MITEs are present in TE-rich regions and may be captured in retrotransposon transcripts.

You report that MITE-Undine is present in gene-poor regions of the Zt genome. This is usually a good proxy for TE-richness in ascomycete genomes. The paper also mentions that LTR-retrotransposons are among the most abundant elements in Zt. LTRs contain intrinsic regulatory elements (i.e. PolII promoters) which drive their own expression and can become novel promoters for nearby genes / ncRNAs.

If these MITEs are nested within or adjacent to the more abundant LTR-elements they might just be present in LTR-derived transcripts.

3) MITEs may be over-represented in the UTRs of stress-induced genes.

MITEs have a preference for inserting within open chromatin. MITEs require the expression of cognate TPases for transposition and TPases from autonomous TIR-element counterparts are often de-repressed under stress conditions. It would follow that MITEs may disproportionately find themselves inserted in the promoters of genes that are incidentally also highly expressed under stress conditions (or other chromatin that is accessible at that time).

In this case, MITEs may appear to be "expressed" as they are transcribed as part of the UTR of stress-induced genes. This is likely the case for mimps in Fusarium spp. where they are common in the 5' UTRs of early effectors BUT have no observed effect on expression themselves.

This might be less likely specifically for MITE-Undine as it seems to be highly "expressed" under all conditions and generally far from neighbouring genes.

4) MITE-Undine elements may retain cis-regulatory functions.

This would be pretty cool. Do you see transcripts initiating within these MITEs rather than just spanning the full MITE? Could this explain why MITE-Undine is highly expressed in Figure S5A-D?

It would be nice to see a MITE-Undine reference sequence included in the manuscript.

On 2019-05-15 20:39:34, user Annette Anderson wrote:

Very interesting paper on a novel DCTN1 splice variant in bipolar disorder. This relates to our recent unpublished finding of a severely affected boy with autism where we have found a predicted pathogenic variant in him and his autistic mother in the same DCTN1 gene. However, in a different region of the gene, in the highly conserved binding region for the dynein intermediate chain. Thus, it is predicted that this likewise will also result in severe autophagy deficits. Could the author comment on our related finding as a significant proportion (25-30%) of autistic patients also have bipolar disorder?

On 2019-05-04 20:28:12, user Andre wrote:

An important paper just released related to autophagy: autophagy deficits causing decreased surface GABA A receptors https://advances.sciencemag.... Further evidence in the importance of autophagy in neurosychiatric conditions.

On 2019-05-15 19:30:18, user Kunal Dutta wrote:

Dear Readers,

In the spirit of this preprint server, we respectfully solicit any questions, comments or thoughts that would assist this line of research. Thank you all, sincerest regards,

Kunal

On 2019-05-15 17:33:18, user Joe Yeong wrote:

We would like to share this good marker to the field as a preprint. Full paper is under review, with more patients, details and completed clinical info. Welcome any comments and validation!

On 2019-05-15 11:05:14, user Guest wrote:

It makes no sense that Tuscans have South Asian admixture, and no other study shows that. Something's wrong. And in Figure S9C, ALL of the European samples show almost as much non-European (mostly South Asian) admixture as the Afrikaners. That makes no sense either.

On 2019-05-15 08:45:58, user M Parker wrote:

Combining structural information with machine learning seems to be a powerful approach for predicting drug resistance, however this work seems to have essentially just replicated work that was published last year (https://doi.org/10.1164/rcc..., and replicated many of the same figures, with no real gain in performance or insight.

On 2019-05-15 06:59:58, user Helena Richardson wrote:

This is a really interesting paper, describing an important mechanism by which glioblastomas develop.

On 2019-05-15 06:05:08, user Helena Richardson wrote:

This is a seminal paper, which has important implications to human glioblastoma prognosis and therapy!

On 2019-05-14 19:11:55, user Gary Brazzell wrote:

We should not attribute the measured gains to respiratory muscle training (RMT) based on this study, but the study does do a nice job of quantifying the gains available through home health.

A previous meta-analysis found that RMT alone was not better than wait list control groups.(1) However, COPD patients receiving lower extremity training only performed better on shortness of breath tests (1), and other studies have found that upper extremity training improves ventilator muscle recruitment.(2-5) Based on this evidence, it would be safer to say that the lower and upper extremity training in this study did more to ameliorate COPD symptoms than the RMT. Future investigations should compare home health with and without RMT to discern the usefulness of RMT. It is likely that shortening the rehab program to only its effective elements can improve adherence.

References:<br /> 1. Salman G, Mosier M, Beasley B, et al. “Rehabilitation for patients with chronic obstructive pulmo-nary disease: meta-analysis of randomized controlled trials.” J Gen Intern Med. 2003: 18 (3): 213-21.<br /> 2. Celli B, Rassulo J, Make B. Dysynchronous breathing during arm but not leg exercise in patients with chronic airflow obstruction. N Engl J Med. 1986; 314: 1485-90.<br /> 3. Criner G, Celli B. Effect of unsupported arm exercise on ventilator muscle recruitment in patients with severe chronic airflow obstruction. Am Rev Respir Dis. 1988; 138: 856-61.<br /> 4. Martinez F, Couser J, Celli B. Respiratory response to arm elevation in patients with chronic air-flow obstruction. Am Rev Respir Dis. 1991; 143: 476-80.<br /> 5. Costi S, Crisafulli E, Antoni F, et al. Effects of unsupported upper extremity exercise training in patients with COPD – a randomized clinical trial. Chest. 2009; 136: 387-395.

On 2019-05-14 17:12:51, user rpwnhlbi wrote:

Not sure promoting private email addresses is worth the risk, even if they're more permanent. Too many examples of fake authors / reviewers involved in the publishing process as a result. ORCiD helps address this pretty well, but barring such verification, I'd always be skeptical of Gmail (etc) addresses as a journal editor or conference reviewer. I would indeed echo the recommendation for institutions to install forwarding services to departed faculty email accounts.

On 2019-05-14 13:06:09, user John Wilson wrote:

Dear Readers,

In the spirit of this preprint server, we respectfully solicit any questions, comments or thoughts that would assist this line of research. Thank you all, sincerest regards,

John

On 2019-05-14 06:26:40, user Ashwani Kumar wrote:

The full article related to above article : <br /> Structural basis of hypoxic gene regulation by the Rv0081 transcription factor of Mycobacterium tuberculosis <br /> has been published with some modification in <br /> FEBS Letters<br /> Volume 593, Issue 9<br /> Pages: 982-995<br /> May 2019<br /> https://doi.org/10.1002/187...<br /> https://febs.onlinelibrary....<br /> https://febs.onlinelibrary....

On 2019-05-14 05:17:04, user Preeti Garai wrote:

This manuscript from Garai et al. has been recently accepted for <br /> publication in PLOS Pathogens. Significant changes have been made to the <br /> preprint during the revision process and a link to the published article <br /> is forthcoming.

On 2019-05-13 20:31:05, user Yang Li wrote:

Is the Supplementary Material missing? I was hoping to find a link to download the Supporting Info document

On 2019-05-13 14:43:41, user Jan Janouškovec wrote:

Hi, this looks quite useful and in line with earlier conclusion based on long Sanger-derived rDNA contigs. A couple very small suggestions about the apicomplexan part, even if this is not quite in the focus. You talk about resolving "neogregarines" as a group with the combined dataset but several papers before you have shown they are polyphyletic. Ascogregarina is not even a neogregarine and you can't really compare the poor sampling in your concatenated tree to Rueckert et al., 2011 or other publications. I'd consider avoiding the name altogether (they are all eugregarines, really). I would also not introduce gregarine as "paraphyletic" since this question is far from resolved. Of note, Simdyanov et al., 2017, Peer J, have done good work on comparing the resolution power of the 18S+28S concatenation over 18S alone in gregarines and all apicomplexans, including some <br /> relationships discussed in this paper; perhaps something to mention here too. Good luck.

On 2019-05-12 11:13:42, user Oliver Pescott wrote:

Our paper on using Bayes rule to assess patterns of co-occurrence between rare and common species might be of interest? https://www.researchgate.ne...

On 2019-05-11 17:36:32, user Luis Mauricio T.R. Lima wrote:

Dear readers,<br /> I've noticed an error at lines 200-201 of the manuscript, where zinc content are inverted between control and intervention<br /> Please read "(38 mg/kg CONTROL vs 11 mg/kg INTERVENTION)". Throughout the manuscript and analytical certification (Supplemental Material) these data are stated correctly.

Sorry for the inconvenience, thanks for comprehension.

Best, Mauricio

On 2019-05-11 15:09:48, user Yan Wong wrote:

Hello, <br /> I am a graduate student from the University of Toronto and I was really amazed by the EWAS approach and findings of the study. I am very interested in reading more about the correlation between blood and brain methylation, but I was unable to find the related supplementary figures in the article. I am, therefore, writing to see if the supplementary figures and tables could be found anywhere on the BioRXiv website, or how I may find access to them? Thank you very much in advance.<br /> Yan

On 2019-05-10 20:03:29, user Yichao Li wrote:

The method in this paper: PWM -> TFM-pvalue -> k-mers -> intersect with dnase-seq -> assign score using chip-seq

I think essentially, this process is given true / false labels defined by chip-seq, build a machine learning model using k-mers and dnase-seq. Then I think a machine learning model would beat the method stated in this paper.

To better predict the effects of SNPs, first, you have to be able to accurately predict transcription factor binding sites (ENCODE-DREAM); otherwise, how can we believe the prediction of SNP effects are true? I think for this paper, it's missing this part.

On 2019-05-10 18:34:19, user Tartaglia Lab wrote:

the link to the algorithm: http://service.tartaglialab... (lower case)

On 2019-05-09 10:37:25, user Tartaglia Lab wrote:

Will RNA have the same structure in vitro and in vivo?

On 2019-05-10 14:49:48, user José María Martín-Olalla wrote:

A screencast describing this manuscript is available in youtube

https://youtu.be/GXx3lcCTkFQ

On 2019-05-10 06:28:34, user Milind Watve wrote:

Our manuscript was rejected by a leading journal with comments by three reviewers. We expressed our desire that in the spirit of transparency of the review process, the reviewers’ comments and our responses should be allowed to be posted and made public. Two of the two reviewers and the journal editors agreed to the request and therefore we are posting their comments and our responses to them here. Although the journal editors consented to post them, on the advice of Biorxiv admin, we are keeping the journal, editors as well as reviewers anonymous. <br /> Rejection is a part of the game and we respect the editors’ decision. However, the reasons for rejection should be transparent so that readers can make their own judgment about the fairness of the editorial process. Transparency would make the review process more responsible and we express our full support to it. <br /> We thank the editors and all the three reviewers for their inputs. We would have been happier if reviewer 1 also agreed to post his comments.<br /> Milind

Reviewer #1:<br /> Did not respond to the request for consent to post the comments.

Reviewer #2:

The authors provide a systematic literature study on the question: “does insulin signaling decide glucose levels in the fasting steady state?”. The answer is a clear no. Although the overview looks solid - I am not an expert in all the literature on glucose homeostasis, so I cannot decide on that, really – the conceptual aspects of this study are rather weak. This may very well reflect the general weakness in conceptual thinking in biomedical sciences, but certainly the control engineers that build feedback control system for artificial pancreas applications will find the answer trivial. The authors use biologically fuzzy terminology, such as “drivers” and navigators”, CSS and TSS, and later r and K strategies, where terminology of control theory would be most appropriate. Not a single reference to control theory, where an integral feedback principle could explain much, if not all of the observations, it seems.

Response: The reviewer appropriately captures the state of control theory and models by the words “much, if not all”. All the models of glucose homeostasis today explain only a small part of the demonstrated features of glucose homeostasis and of diabetes. The “much” is a very small fraction of reality and most models stop at explaining only some of the features. Not being able to explain a certain empirical finding does not immediately invalidate a model. However, a direct contradiction with empirical findings certainly raises questions about the model. The model suggested by the reviewer below is an excellent example of it.

For illustration: if the CSS model that the authors use in the supplements is slightly modified by:

dGlc/dt = (Gt+L) – K1 Glc – Ins_sens K2 ins<br /> dIns/dt = K3 Glc - d

(so insulin removal is independent of the insulin level), then at steady state of this coupled system (where dGlc/dt = dIns/dt = 0):<br /> Glc_s = d/K3<br /> Ins_s = {(Gt+L) – K1/K3 d }/(Ins_sens K2)

Thus, Glc at steady state is independent of insulin sensitivity, or glucose production or consumption. It is also said to be perfectly adapted to these parameters. So if Ins_sens is lower, Ins_s will be higher but glc_s remains the same: a perfect basis for the HOMA index!<br /> Only the experiments with reduced removal of Ins (parameter d) would be expected to have lower glucose, but of course this is a very very simple model of glucose homeostasis. Also poor synthesis of insulin by impaired beta cells would lower K3 and this may explain higher fasting glucose levels.

Response: This is an interesting model and a perfect example of how in order to explain one empirical finding the model contradicts many others. Certainly the model accounts for hyperinsulinemia in response to insulin resistance without a change in glucose level. However, it does not explain the results of insulin degrading enzyme knockouts, which would decrease d and is thereby expected to increase glucose, but that does not happen in experiments. Further we simulated using this model to see whether the FG-FI correlation in the steady state would be different than during post glucose load dynamics. Even in this model the regression correlation parameters remain the same and only the range shifts upwards. Thus the model suggested by the reviewer does not account for the experimental and epidemiological results that we cite in this manuscript. <br /> The focus of our manuscript is to look at convergence of many sets of experiments and therefore suggesting a model that satisfies one but not others is not an appropriate solution. <br /> The other problem with the model suggested by the reviewer is that it makes an assumption of constant degradation rate of insulin independent of its standing concentration. Most biochemical decays are known to follow negative exponential. If you want to make an assumption deviant with the general pattern, you need a justification and validation for the assumption. In the case of insulin there is published literature on the half-life of insulin.So the baseline assumption should be that insulin degradation follows half-life dynamics and if you want to make any other assumption, you need convincing justification for it.<br /> So I am a bit puzzled. What is the point of this paper? Does anyone take CSS seriously, really? Again, I do not know all the literature but I am sure there are good models out there that can and do explain T2D and glucose homeostasis very well. <br /> Response: The whole point is that in existing there isn’t a model that does so. Believing that there are good models out there is not sufficient for the reviewer. If there is any kindly point it out specifically. <br /> Should ….(Journal name)…. fix a failure in the education of doctors? And if ….(journal name)… decide they want to do that, please teach them the right vocabulary and conceptual frame work, and properly cite the control theory literature!<br /> Response: We would be glad if control theory has a model that is compatible with all the empirical results pointed out in our manuscript. It is not enough for the reviewer to say that there are. Kindly point out specifically if there really are. As far as we know there aren’t any. But this manuscript is not an intended review of models, it rather lays out the set of experimental results and epidemiological patterns that any model of glucose homeostasis needs to explain. This set has been put together for the first time and that is the main contribution of the paper. Our central argument is that glucose homeostasis needs to take into account all these results TOGETHER. You cannot look at partial picture again and say there are models that are compatible with the partial picture. <br /> To the best of our knowledge, none of the existing models would explain all of them together. We are suggesting here that this is because the set of foundational assumptions of these models is not correct. We are suggesting what change might be needed in it. Building models with the new set of assumptions would certainly deserve a separate publication. Our manuscript is not intended to give the answer, we are defining the question in a broader perspective that has not been taken so far.

Specific comments:<br /> 1. “The belief that this product (HOMA) reflects insulin resistance is necessarily based on the assumption that insulin signalling alone quantitatively determines glucose level in a fasting steady state.”<br /> I really do not get this. See the above simple model: many parameters determine the steady state levels, but if Ins_sens is lower (or L is higher by less insulin inhibition), steady state insulin is higher at the same glucose concentration, so HOMA makes perfect sense to me. Obviously, there can be other ways to change HOMA, but it is simple and effective in the clinic.<br /> Response: HOMA does make sense w.r.t the above model but as pointed out earlier this model has multiple flaws and unless we have a model that is compatible with all experimental and epidemiological results it is difficult to claim that HOMA makes sense.

1. “There is a subtle circularity in the working definition of insulin resistance. Insulin resistance is blamed for the failure of normal or elevated levels of insulin to regulate glucose…. However, clinically insulin resistance is measured by the inability of insulin to regulate glucose. Such a measure cannot be used to test the hypothesis that insulin resistance leads to the failure of insulin to regulate glucose.”<br /> Sorry but the circularity is so subtle that I miss it. If the argument is that insulin regulation is impaired in insulin resistance (what’s in the name), people should measure the action of insulin, right? What is wrong here?<br /> Response: To explain the circularity in different words-<br /> (i) Insulin is unable to regulate glucose because the body has insulin resistance<br /> (ii) Insulin resistance is measured as the inability of insulin to regulate glucose<br /> (iii) Put (i) and (ii) together, it reads “insulin is unable to regulate glucose because of the inability of insulin to regulate glucose”<br /> Isn’t this circular enough or is more clarification needed?

2. line 437: suddenly, “hysteresis” appears out of nowhere. What is this? Please explain properly if relevant, do you really think these poor doctors know what that is?<br /> Response: We agree and will revise the text here to explain the context without the word “hysteresis”.<br /> In brief, the comments by this reviewer are thought provoking and we learnt a lot while addressing them, but they leave us with a little bit of doubt about the soundness of his/her ideas about control theory. <br /> --

Reviewer #3:

This is a very interesting question, and a novel approach to addressing it. I have focussed primarily on the systematic review aspects.<br /> 1. The meta-analysis technique used is essentially "vote counting", and this is not recommended (see https://handbook-5-1.cochra... for reasons given in the reference.<br /> Response: Many many thanks to the reviewer for pointing this out. We read the link carefully to find that our analysis is very sound by these guidelines. It does not recommend vote counting in significant versus non-significant types of outcomes. But it clearly says, <br /> “To undertake vote counting properly the number of studies showing harm should be compared with the number showing benefit, regardless of the statistical significance or size of their results. A sign test can be used to assess the significance of evidence for the existence of an effect in either direction”<br /> This is precisely what we have done. So this comment validates our analysis and increases our confidence. Thanks once again. <br /> 2. I could find no mention of a PROSPERO registration - this is important<br /> Response: We agree and will improve during revision.<br /> 3. There is no attempt, as far as I can see, to address the possibility of publication bias<br /> Response: Publication biases are discussed already in the main text line 125-129, but we will elaborate more and also include in supplemental table 3.<br /> 4. The analysis is not reported in a way consistent with the PRISMA guidelines (although these relate to reviews of human data, they have lessons for animal reviews<br /> Response: We made our best attempts to follow PRISMA guidelines for animal experiment reviews as well. It would have been more useful if any inconsistency was specifically pointed out by the reviewer.<br /> 5. There is, as far as I can see, no assessment of risks of bias in the contributing animal studies<br /> Response: We agree and would be glad to improve on. <br /> 6. In my view, it is not enough to say that data will be made available on acceptance - part of peer review should be to ensure that it is made available in a form which is complete, comprehensible and useable, so it needs to be avaialble (even if only through a private link) at this stage.<br /> Response: That is certainly possible and will be done for the revised version.

Regarding the animal experiments these should be reported according to the ARRIVE guidelines, and as far as I can see (I may have missed it, or you may have done it but not reported it) these were non randomised unblinded experiments without an a priori sample size calculation.<br /> Response: We see the importance of reporting these details for the primary experiments that we performed, but for the review and meta-analysis section we do not have control over what the authors did.<br /> In a nutshell, comments by all the three reviewers are a convincing reinforcement that our central argument is sound and strong. We agree with many of the refinement suggestions and look forward to publish a revised version soon.

On 2019-05-09 20:06:00, user Derek S Lundberg wrote:

Looks very interesting. From a quick glance at figures before reading, want to mention:

Pls call it Bray-Curtis dissimilarity - its not a true "distance". https://en.wikipedia.org/wi...

The Shannon Diversity Index is what is graphed, and it's better to label it as such (the Index). Shannon Diversity is the exponential of that graphed, pls see: http://www.loujost.com/Stat...

On 2019-05-09 18:18:43, user Tony Warne wrote:

New version now available: https://science.sciencemag....<br /> Sorry it took so long!

On 2019-05-09 15:15:48, user Amit Singh wrote:

New mechanism underlying HIV-TB synergy.

On 2019-05-09 13:51:49, user David Hoksza wrote:

Very cool. But the overview of existing tools might might be missing ProtVista and more importantly MolArt, the tool integrating ProtVista and LiteMol tools into a single javascript plugin.

On 2019-05-09 09:59:17, user Koen van den Dries wrote:

The cartoons in Figure 1B and Supplementary Figure 1 seem to incorrectly depict the myosin IIB tension sensors (except for the S1 variant). The myosin IIB motor complex is composed of two myosin heavy chains which is why two modules should have been drawn if I am not mistaken. Same is true for the depicted point mutation and the N-terminal mTFP (which should be present at both heads in the dimer).

On 2019-05-09 02:59:04, user Yanmei Huang wrote:

Hi, Thanks for creating a very nice tool! One question: what is the definition of "Noise" and "Signal" in Figure 2 and Figure 3? How are they calculated?

On 2019-05-08 11:43:44, user FEGraf wrote:

Where can I access the supplementary data?

On 2019-05-07 16:55:51, user Author update wrote:

The crystal structure of Nocturnin with NADPH is now released: https://www.rcsb.org/struct...<br /> Deposited: December 18, 2018

On 2019-05-07 15:00:27, user Jacob Robino wrote:

I'd *really* like to know whether the simulation assumed whole-genome sequencing or a simple SNP chip.

On 2019-05-07 14:12:56, user Zaved Hazarika wrote:

Dear Marziyeh<br /> I have performed a similar work. However in our MD of ZnO NP in water (4nm,2838 atoms, Amber99sb ff), all same parameters considered, the ZnO NP is not getting stabilized over 100ns time. Can you kindly provide your raw data to cross check with ours?<br /> Regards<br /> Zaved Hazarika<br /> Research Scholar<br /> Dept.of MBBT,<br /> Tezpur University<br /> India<br /> email:zaved@tezu.ernet.in

On 2019-05-07 13:19:06, user 刘科辉 wrote:

Could gencore build consensus sequence with umi from PacBio Sequel reads？

On 2019-05-07 09:32:09, user Daniel Blaese wrote:

I'm not trying to be pedantic, but shouldn't the title be corrected to "Carbon monoxide dehydrogenases enhance bacterial survival by oxidizing atmospheric CO"? The organism of interest doesn't respire the CO either, it respires oxygen. It is obvious that the authors understand this difference (2nd sentence of the abstract) but I find it very strange to say the organism respires CO when it does not respire CO. Just like E. coli does not respire glucose during aerobic growth. Or am I missing something here?

On 2019-05-07 01:17:57, user Keith Robison wrote:

I've written a pair of analyses on this -- some key criticisms are the lack of technical detail and that the analysis of errors is much less detailed than desired. There's also the lack of specificity in the text as to which data was deposited publicly -- only the E.coli is available. Also, the phred quality scores are overestimated at the high end by perhaps as many as 5 points.<br /> Poking at Genapsys Preprint<br /> Genapsys' Base Caller: Mysterious, But Not Ideal?

On 2019-05-05 09:31:16, user David Curtis wrote:

I looked at this for schizophrenia with negative results:<br /> https://journals.lww.com/ps...

I don't know whether this was one of the traits you tested? It might be worth discussing why your results differ from mine?

On 2019-05-04 21:59:22, user darachm wrote:

Interesting.<br /> Typo on line 72.<br /> On line 142, should "application" be "amplification"?

On 2019-05-04 18:37:34, user L. T. Fang wrote:

Detailed documentation of the somatic mutation analysis can also be found at https://sites.google.com/vi.... That page is continuously maintained and updated to reflect the latest finding and version updates with regard to the somatic mutation "ground truth."<br /> Comments are welcome here.

On 2019-05-04 04:53:24, user micknudsen wrote:

Where does one find the data? I have tried searching for SRP162370 at https://www.ncbi.nlm.nih.go..., but there are zero hits.

On 2019-05-02 17:51:40, user Alexander Kozik wrote:

Lactuca mitochondrial genome sequences were publicly released at NCBI GenBank

https://www.ncbi.nlm.nih.go...<br /> https://www.ncbi.nlm.nih.go...<br /> https://www.ncbi.nlm.nih.go...

On 2019-05-02 09:03:06, user david morris wrote:

"One is that you quote the AUC for only the best performing GRS. It ought to be obvious that you can't just try a number of different ways of getting the GRS and then only highlight the one which does best. (Especially as you applied it to three different datasets and only report the best of the three.)"<br /> REPLY: We highlight in the paper that a GRS using SNPs that pass p < 1E-05 consistently does best across all two-way training and test data sets. Few studies have compared this for the same disease in multiple GWAS. We are open about these tests in the paper and the take away message for this point was the likely polygenetic nature of SLE, however this was not the main result or the main point which was concerned with the GRS loading in renal patients.

"Another concern is that you don't say how much the GRS improves the AUC compared with a predictor just derived from HLA."<br /> REPLY: The HLA or more widely the MHC does not do well when used as a region to form a GRS due to it being very difficult to derive a good model of association that agrees across multiple data. This is due to it not being well typed in terms of structural variation. In short it would be naïve to use the MHC in GRS without better coverage of the locus.

"Also, in the methods section you suggest that the effect size for each SNP is derived from its original publication but subsequently it looks as though you have fitted your own effect sizes using a training set."<br /> REPLY: This is true for the GRS used in renal prediction (only published sle associated SNPs). We estimated effect sizes in training sets when looking at the GRS across multiple GWAS for SLE.

"There are huge ancestry effects in SLE. How do you know you haven't just picked up a GRS for ancestry? (https://www.ncbi.nlm.nih.go... plus lots of other recent papers) "GRS has been showed to be predictive for several diseases including cardiovascular disease (AUC=0.81, 95%CI: 0.81-0.81)" - Um no, that is not the AUC for the GRS, as you point out later."<br /> REPLY: Agreed on last point, we need to clarify our take on that paper. This helps our argument that GRS for SLE does relatively well. The point on ancestry: the paper you refer to gives several reasons for the observation in Schizophrenia of the mean PRS differing between African and Europeans including that the PRS does indicate susceptibility to disease and that observed differences are due to negative selection, but does not have any evidence to settle on one explanation and like other papers on this issue is more a warning to consider these points, which we agree with, rather than dismiss results on PRS.

"Finally, you suggest that the GRS might "assist early prediction of lupus nephritis in a clinical setting". I strongly doubt this is the case. What is the magnitude of this effect? What would be the clinical utility of such a predictor? (See https://www.nature.com/arti..."<br /> REPLY: Why do you strongly doubt this? We have found that the SLE GRS does segregate SLE Renal from SLE non-renal. We do not suggest this is ready for clinic now but that if this finding is real then better powered GWAS and clinical studies may be enlightening. The clinical utility should be obvious as renal SLE is a serious conditional.

On 2019-04-28 15:26:27, user David Curtis wrote:

I have a number of concerns about the AUC you quote. One is that you quote the AUC for only the best performing GRS. It ought to be obvious that you can't just try a number of different ways of getting the GRS and then only highlight the one which does best. (Especially as you applied it to three different datasets and only report the best of the three.) Another concern is that you don't say how much the GRS improves the AUC compared with a predictor just derived from HLA. Also, in the methods section you suggest that the effect size for each SNP is derived from its original publication but subsequently it looks as though you have fitted your own effect sizes using a training set. There are huge ancestry effects in SLE. How do you know you haven't just picked up a GRS for ancestry? (https://www.ncbi.nlm.nih.go... plus lots of other recent papers) "GRS has been showed to be predictive for several diseases including cardiovascular disease (AUC=0.81, 95%CI: 0.81-0.81)" - Um no, that is not the AUC for the GRS, as you point out later. Finally, you suggest that the GRS might "assist early prediction of lupus nephritis in a clinical setting". I strongly doubt this is the case. What is the magnitude of this effect? What would be the clinical utility of such a predictor? (See https://www.nature.com/arti...

On 2019-05-01 21:54:01, user Brian DeVeale wrote:

Awesome work! Perhaps the wrong forum, but it looks like the command for converting v2 objects to v3 is 'UpdateSeuratObject' in R and listed as 'UpgradeSeuratObject' in the FAQ on your website.

On 2019-05-01 16:07:38, user Bert wrote:

Interesting, but there is no evidence yet that there is actually a TonB-dependent (outer membrane) transporter that imports these compounds.

On 2019-05-01 11:53:51, user Manas Chakraborty wrote:

This paper is now published at Nature Communications. Please follow the link below. <br /> https://doi.org/10.1038/s41...

On 2019-04-30 16:23:59, user JSRosenblum wrote:

It's nice to see that the authors of this paper used our ERK5 inhibitor, AX15836. However, the conclusion they drew from its use do not agree with what I'd conclude. Here's what they wrote: "While AX15836 had no anti-proliferative effect on two-dimensional cultures over a short period (48 h), when cells were<br /> grown in 3-D cultures (colony forming assay), the IC50 of AX15836 on A375/211 cells was<br /> reduced from 62 uM to 6.7 uM, indicating that A375/211 cells were more sensitive to drug than A375 cells (Fig. 4D). "

The compound is a low nM inhibitor of ERK5. What it is doing over the course of 2 days in 3D culture at 1000-times its ERK5 IC50 is quite likely to be derived from something other than ERK5 inhibition. The 9-fold increase in potency in these cells in 3D versus 2D culture is interesting, but likely not ERK5-derived.

Finally, I'd suggest they see what a pure bromodomain inhibitor does in their system, for example JQ1.

On 2019-04-29 17:55:26, user Charles Warden wrote:

I hope we can find a way to get more comments on bioXriv, so that they could be discussed prior to submission to a peer reviewed-journal:

https://www.nature.com/arti...

I would much rather have a revised pre-print than a correction / retraction in a peer-reviewed paper.

You can see any comments for an individual from their Disqus account, but I think this worked well in terms of keeping a friendly tone and asking questions that I think could improve reader understanding: https://www.biorxiv.org/con...

On 2019-04-29 15:01:23, user xbdr86 wrote:

Paper already published.

Bofill-De Ros X, Kasprzak WK, Bhandari Y, Fan L, Cavanaugh Q, Jiang M, Dai L, <br /> Yang A, Shao TJ, Shapiro BA, Wang YX, Gu S. Structural Differences between<br /> Pri-miRNA Paralogs Promote Alternative Drosha Cleavage and Expand Target<br /> Repertoires. Cell Rep. 2019 Jan 8;26(2):447-459.e4. doi:<br /> 10.1016/j.celrep.2018.12.054. PubMed PMID: 30625327; PubMed Central PMCID:<br /> PMC6369706.

https://www.cell.com/cell-reports/pdfExtended/S2211-1247(18)31984-3

On 2019-04-28 23:28:47, user Danielle wrote:

Published version of the paper now available from Plos One

https://doi.org/10.1371/jou...

On 2019-04-27 15:04:45, user Paul Schanda wrote:

This publication has been accepted in Nature Communications.

On 2019-04-26 22:15:24, user Joylynn Woodruff wrote:

I'm curious as to the names of the specific birds that are migratory and threatened birds. The migratory path should include at least the area north of the Dallas/Ft. Worth area.

On 2019-04-26 17:10:06, user Kristen Naegle wrote:

From the UVA Systems Biology Journal club discussion of this paper 4/23/19

We found this to be a really interesting paper with a timely machine learning method on a topic with a lot of room to advance. The authors do a great job motivating the needs in the field, based on limitations of existing methods. Specifically, it is exciting to see a method that seeks to learn globally from all kinases and to extract kinase features that shape kinase-substrate specificity. We found we could not completely understand some key features of the model and its use with the text as it stands and we hope our experience with this manuscript, as outlined below, will be of help to the authors.

Models and model interpretation<br /> We had some confusion about the model as implemented, especially around whether certain aspects were used to make the model interpretable vs. what was in the model. <br /> 1. PSSMs: A major strength of the neural network approach is the ability to learn and encode conditional dependence between positions in the kinase and amongst positions in the substrate. However, as currently depicted in the approach, it seems that the final predictor relies on collapsing the RNN model into a PSSM and scoring substrates across RNN-derived PSSMs. If this is the case, it is unfortunate to rely on a scoring methodology that is incapable of incorporating conditional dependence between positions. It would be great if the paper could clarify the methodology and explore prediction results that avoid the PSSM as a primary scoring function. <br /> 2. Attention Matrix: The attention matrix is really interesting and has a lot of power to explore specificity determining positions. However, we were unclear about some of the details about the attention matrix, its use, and its presentation in this work:<br /> 2a. Is the feature selection process that determined the attention matrix values used in the final classifier? As written, we were unclear about this. On the one hand, performance as a function of forward feature selection was given. On the other hand, if there are ultimately only 15 kinase sequence features used, then it seems unlikely that that broad range of mutations lands in those features and would make it impossible to score differences as a result of kinase mutations. <br /> 2b. The attention matrix in Figure 2 appears to highlight more than 15 kinase features, and suggests there are family-specific kinase features. However, the text suggests there was a universal set of 15 kinase features. How these 15 were chosen was also under debate in terms of the effectiveness and resolution of the feature selection method. Given the intense growth in performance between 5 and 15 features, it seems it would be beneficial to increase the testing of performance at a higher resolution (1:15 features with one at a time addition).<br /> 2c. It was clearly stated how many features selected by DeepSignal overlapped with KinSpect and DoS, but it would also be nice to know how many KinSpect and DoS features were not identified by DeepSignal (set differences vs. set intersections). <br /> 3. Model Details: <br /> 3a. Is this a “deep” neural network - where are the layers of convolution? Are there hidden layers?<br /> 3b. What are the exact inputs to the model?<br /> 3c. How long is the sequence retained in the recurrent neural network? Is there a limit to how far back the LSTM considers? <br /> 3d. How is allostery incorporated in the model (e.g. as conditional dependence)? Long-range interactions not encoded in local sequence space would appear to be missed unless the entire sequence is considered throughout the recurrent neural network.

Figure 3 and related methods:<br /> The choice of negative data is hard when the training set only contains positives. The authors used a method that is consistently used in the field. However, because it is a random draw and makes many assumptions about the draw (that there are not false negatives in the set), we felt it would be beneficial to test the robustness of conclusions drawn by repeating this analysis across many resamples of a negative set. This would help us understand the sensitivity or robustness of the conclusions to that particular selection of data. Additionally, it is not clear what model hyperparameters have been tuned to generate the precision-recall and AUROC analyses for the comparator predictors.

Generalizability of learning on global kinases and training misbalance<br /> We were intrigued by the results in Figure 2E. We think this is a really interesting experiment to test applicability of a globally learned model. We noticed that the only tyrosine kinase in this batch (as a result we assume of being the only tyrosine kinase with more than 100 substrates annotated in the training set) was affected the most when predicted by a model of all kinases in that set, when compared to a single-kinase SRC model. We feel that may suggest that if a training set is predominantly skewed towards serine/threonine kinases that it will not produce the ideal model for tyrosine kinases. As tyrosine and serine/threonine signaling are separated both evolutionarily and physicochemically, it seems reasonable to make two models of kinase-substrate predictions and explore the results of those independently to assess whether the attention value matrices and performance differ greatly. We also wondered if data skew in Figure 2E analyses or more broadly could be a factor (perhaps it would be beneficial to add an analysis of the training data itself).

Mutation analysis<br /> In addition to the confusion we noted earlier about how the attention value matrix and feature selection is wrapped back into the model and its effect on the ability to test mutational effects, we also wondered what the “false positive rate” was on determination of cancer genes as a function of kinase-substrate misregulation using DeepSignal. The authors focus on capturing known oncogenes (as a function of percent covered), but we wished to know how many total were predicted to be detrimental and whether this differed greatly between DeepSignal and MSM/D-PEM (i.e. both specificity and sensitivity). One representation that might be helpful is to display the total number of predicted cancer genes with the proportion of true highlighted in the subset.

SH2 domain analysis<br /> As some of our members are very familiar with the problems with the published SH2 domain data (e.g. that they cannot be merged as there are disagreements, different types, and different scales), we understand why the authors chose to build individual models for each dataset. However, in the mutation analysis, it is unclear what final SH2 domain model they used and the authors do not provide the same level of detail on what was learned in the SH2 domain as they did for kinases. In addition to providing more clarity in the methods used for mutation analysis (as it relates to SH2 domains), it would likely be beneficial to do a sensitivity analysis in the outcomes about predicted oncogenic mutations as a result of isolating the kinase and SH2 domain components. Finally, although the paper used was cited, it would be helpful to describe in more detail exactly how an oncogene was determined for readers to better interpret the method and results provided here.

Signed by:<br /> Kristen Naegle, Ben Jordan, Kevin Janes on behalf of the University of Virginia Systems Biology Journal Club (Journal Club of 4/23/19)

On 2019-04-25 22:02:20, user Julio Retamales wrote:

Good paper! A detail: in the second paragraph of the abstract it should read: "The green revolution gene". The "r" letter is lacking.... <br /> Thanks!

On 2019-04-25 19:30:54, user Madhavi Adiga wrote:

Hi, <br /> I'm Madhavi Adiga, just started my graduation in Pharmacology Dept. I'm interested in Tumour angiogenesis and I want to build my project in this field. As for the starter I proceeded with different tumor model system. While across searching I found this article, in which LLC subcutaneous tumor model system is being used. Several other studies shows injecting LLC cells with matrigel as substrate to minimize the variability in tumor size during the course of tumor growth. My question is, you started with low number of LLC cells to inject with and studied up to 16 days without using any solid substrate support as there may be chances of leakage into surrounding tissues rather than being confined to the injected place (as I think this may lead to variability between the groups we study), how this will be different from being used with a solid substrate (matrigel) as the solid substrate may give more support to tumor cells to grow in a confined region. Secondly, on what basis you took 16days as criteria to sacrifice mice? based on humane endpoint criteria? or did you do any growth curve study to select 16days? If you keep more time the knockout mice you used (s1pr1) develop more tumor? <br /> Please let me know as this may be helpful for me for my further studies.

On 2019-04-25 15:28:19, user Pedro Madrigal wrote:

Hi,

Thank you very much for posting the ChIA-DropBox preprint. Unfortunately, it looks like the GitHub repositories mentioned in the preprint (see below) are not available.

Best regards,<br /> Pedro

Code availability<br /> The ChIA-DropBox data processing pipeline is publicly available at: https://github.com/TheJacks.... The ChIA-view visualization tool for multiplex interactions is publicly available at: https://github.com/TheJacks.... Both repositories contain README files with further details on how to get started using the software.

On 2019-04-25 15:13:46, user John Lowrie wrote:

Very good that there are studies to confirm the everyday findings of human rights workers over many years. The report does not venture in to the hows and whys nor what needs to be done to protect these young women (and by the way some boys and transsexuals). Families and local communities could and should do more, especially by facing up to the realities that all work away from home is likely to be risky, and not simply expect and take their hard-earned money. A recent study for Future Forum highlighted that such remittances may not be used wisely. (https://docs.voanews.eu/en-...

There is a common deeply-embeded belief that older children, especially daughters, must make sacrifices for their families. (http://anorthumbrianabroad....:JtembqSkHCPXOa5FCUSSy5rFsLM "http://anorthumbrianabroad.blogspot.com/2016/12/more-to-peering-than-appears.html)")

Beyond families and communities responsible action is needed from authorities instead of their active involvement in promoting the entertainment business. It is a fact that formal and informal permits and documents are needed and issued, and this is a lucrative form of income, thus encouraging rather than deterring the business. Quite often I have observed beer garden owners having to supply food and drink "on-the-house" to government officials as part of the deal.

Finally the international beer companies own drinks companies or have interests in production and supply. Some advocate "responsible drinking" on product labels. None apply the same message for the workers who market them. They abrogate responsibility.

On 2019-04-25 14:55:47, user Jo Wolfe wrote:

The peer-reviewed paper is now published in Proceedings B: https://royalsocietypublish...

On 2019-04-25 10:15:03, user Adam Selwith wrote:

"we provide the phased 10X variant calls as supplemental data"<br /> You may say so, but there are two supplemental data files and both are about reads, not variants.

On 2019-04-25 08:30:20, user °christoph wrote:

this beats phage-mediated HGT by lengths, so to say! and just think of the amount of nucleotides the up-taking cell gets from the degraded strand. interesting study!

On 2019-04-25 07:50:38, user Axel Thielscher wrote:

The results shown in this paper should be contrasted with the findings published in: https://doi.org/10.1101/611962. There, the value and the limitations of intracranial recordings for validating electric field modeling for transcranial brain stimulation are also investigated in more detail.

On 2019-04-23 09:00:52, user Axel Thielscher wrote:

The results shown in this paper should be contrasted with the findings published in: https://doi.org/10.1101/611962.<br /> There, the value and the limitations of intracranial recordings for <br /> validating electric field modeling for transcranial brain stimulation <br /> are also investigated in more detail.

On 2019-04-25 04:19:35, user Jessica wrote:

The links to access the software and data don't work

On 2019-04-24 23:53:33, user Andrew Morgenthaler wrote:

Very cool findings. I think Fig. 1 and Fig. 4 are particularly great, clearly demonstrating the effect synonymous mutations have on fitness and, as a result, on evolution. I wish there was a bit more convincing experimental evidence of the mechanisms by which synonymous mutations are increasing fitness. Why not do RT-qPCR to clearly show which genes in the operon are increasing their mRNA as a result of the mutation? Can you actually show using 5'-RACE that a new promoter is forming?

As a side note, I have no idea what you mean by a 39-3T mutation. Does that mean codon 39, third position in the codon? It would be nice to have your nomenclature explained in the text.

On 2019-04-24 23:47:23, user Andrew Morgenthaler wrote:

page 3, line 3 should say "...in the non-synonymous set that presumably produce truncated, non-functional protein." (remove the "a" before "produce")

On 2019-04-24 17:21:50, user Enkelejda Miho wrote:

Dear David,

thank you for your comment (one year ago?), somehow this has gone unseen until now. It is great that you took the time to critically read this manuscript and give feedback. Your comments are interesting, even if some obviously address on-going and open discussions of the entire immunology community (e.g., CDR3 relevance vs. length, repertoire holes, biological stochasticity in B cell receptor generation, parallelism T and B cell repertoires).

On the other hand, network parameters here are novel and the observations hold the potential to open further questions and prospectives, a further step in understanding and considering additional aspects. One specific side note: what we know through networks does help to plan/design synthetic repertoires following the observed distributions in CDR3 similarities as the simulation algorithm can be designed to satisfy the conditions.

Thank you genuinely for the discussion.

Enkelejda

On 2019-04-24 11:57:00, user daniele marinazzo wrote:

Thanks!

My comments here

https://pubpeer.com/publica...

and pasted below

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

(partial) review for the preprint.

Disclaimer - COI statement: Matteo Fraschini invited me as guest professor for two weeks in Cagliari in the summer of 2017, and we are regularly in contact, even if no collaboration is ongoing at the moment.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Dear colleagues,

thanks for sharing this work. Here follow some ideas and comments.

Let's start with a conceptual issue: from the evidence that PSD and network metrics are very much correlated, you " conclude that it may represent good practice to report the findings from the two approaches in conjunction to have a more comprehensive view of the results." One might argue that this should be even ore true for two methods that are not connected one to the other, rather than for two methods which mostly reflect the same thing.

Now, one might want to argue why the two measures are correlated, and whether this is specific to the brain. I will provide some indications below.

I will go through two steps:

PLV and PSD are correlated (as also observed here A comparison of spectral magnitude and phase-locking value analyses of the frequency-following response to complex tones)<br /> PLV and network measures are correlated.<br /> Let's start with 1. As clearly stated in Phase locking value revisited: teaching new tricks to an old dog, by Bruña, Maestú, and Pereda " PLV in a given band is related to the average coherence in the frequency range encompassed by the band weighted by their relative amplitude ". In particular they propose a computationally efficient, but more importantly illuminating way to define PLV without using the Hilbert transform and phase extraction, coded in Matlab as

[nc, ns, nt] = size(data); ndat = data ./ abs(data); plv = zeros(nc, nc, nt); for t = 1: nt     plv(:,:, t) = abs(ndat(:,:, t) * ndat(:,:, t)') / ns; end

which makes the relation between PSD and PLV evident. From the code that I pull requested to your repository

https://github.com/danielem...

Now, you used unthresholded, weighted, undirected networks and compute Strength, Clustering Coefficient, and Betweenness Centrality on them.

Even taking a random matrix as "network" it's evident that higher values of the entry correlate with Strength and Clustering Coefficient

On the other hand the Betweenness Centrality is constant.

This mainly reflects your finding, except from Betweenness Centrality, which is where one might argue that the nonrandom structure of the functional network is slightly more modulated.

So, to summarize, the results reported on real EEG data reflect this simple associations, and that's a conforting confirmation.

Then one might wonder how to interpret network measures when the links are statistical dependencies across brain areas. This is of course a topic which is much bigger than a paper :). But indeed the link with a more interpretable local measure is helpful.

For the moment you could expand on what this association can tell us, and if we need network measures of this kind.

Thanks!

On 2019-04-24 10:16:54, user Dan B wrote:

Hi - this example may be of interest for this manuscript:<br /> https://www.biorxiv.org/con...

On 2019-04-24 09:07:25, user Johannes Soeding wrote:

The parameters for the program were optimized on the same 54 genomes that were also used as test set. This is very problematic as it can give too optimistic results. You need a test set that is independent of the set used for optimisation.

On 2019-04-24 07:25:00, user Ashish Kumar wrote:

i wonder what is in between ITGA2 and ACLY .... what connects them ....,?<br /> what are the other metabolic pathways it activates on the expense of ITGA2 downregulation ?

On 2019-04-24 07:15:16, user ST Sebastian wrote:

The topic “theranostic potential” is very misleading to reviewers and readers. Theranostics means therapeutics plus diagnostics. The studies did not show any diagnostic function. Using MRI to monitor drug release from cornea implant is not practical.

On 2019-04-23 20:08:21, user Andrew Whitehead wrote:

I'm curious. Are there data indicating that these morphologies do/not emerge from habitat (e.g., temperature) -induced developmental plasticity? That is, are they heritable? Some of the studies cited (e.g., Sfakianakis et al) show that temperature induces alternate developmental trajectories. The article frames the objectives/data/conclusions within the context of evolution. Perhaps I'm missing something, but maybe the authors could consider whether some qualifying statements about the potential influence of temperature-induced developmental plasticity would be warranted?

On 2019-04-23 16:25:12, user Carstenn wrote:

Negavitve calories are so far I know, meant to be (meat)proteins which demands much energy to digest and which may be why some animals like the wolverine can eat so much compared to others who eat s more energyeconomic foods?

LIzards are coldblooded, right? So they do not need energy to warm up as humans does, and therefore need warmt for digestion which is not be compared to humans, meaning, that humans need much more energy for digestion than lizards does, right?

How then can this study on lizards be compared to humans eventhough their digestion may be similar to ours if their dependence on heat externally supplied is not?

On 2019-04-23 14:47:01, user Falk wrote:

What about retinanet: https://arxiv.org/abs/1708....

On 2019-04-23 07:01:05, user rahul saini wrote:

Hey there,

I want to plot SWNE on my custom dataset(Tabular data).<br /> Please let me know how to proceed?

Thanks in advance

On 2019-04-23 06:43:11, user Michael Alexanian wrote:

Very nice study that highlights, in line with previous literature, the fascinating concept of developmental competence being primed in pluripotency before cell fate decision. We came up to similar observations characterizing Meteor, a pluripotency-specific enhancer that controls mesendoderm specification (Alexanian et al, 2017 - "A transcribed enhancer dictates mesendoderm specification in pluripotency"). The authors could discuss this study in the discussion when they mention two other important studies (Wang et al., 2015 and Kim et al., 2018).

On 2019-04-23 04:08:55, user Protothelenella corrosa wrote:

The foundation of this work is the fragmentation of these<br /> intact glycopeptides to detect informative product ions for confirmation of<br /> peptide sequence, glycosite and glycan composition. If these product ions are<br /> incorrectly annotated, they may provide false confidence of these glycopeptide<br /> characteristics. Accompanying this risk is the relatively niche technique of<br /> AI-ETD, which generates mass spectra containing many (>100) product ions.<br /> Certain assumptions, such as type of fragmentation mechanisms present (in this<br /> paper: b, c, y and z), allow the assignment of these product ions to<br /> glycopeptide fragments.

In figure 1 A, the authors provide a panel of a glycopeptide<br /> MS2 mass spectrum, demonstrating 100% peptide sequence coverage. Examining the<br /> product ion spectrum for this single annotation, we observe over 1000 m/z<br /> values representing product ions ranging from m/z 144 to 1995, confirming that<br /> this spectrum is rich. However, this number of m/z values does not accurately<br /> reflect the true number of detected product ions due to the presence of<br /> isotopes. These can confound the correct assignment of the previously mentioned<br /> peptide backbone and glycan fragments.

If we interrogate the same spectrum in Byonic using the data<br /> analysis files provided by the authors, there are multiple incorrect<br /> assignments of glycopeptide fragment ions to isotopic peaks instead of the<br /> monoisotopic peak. For the spectrum shown in Figure 1A, we found 15/75 (20%) of<br /> the product ion annotations to be incorrect due to this error (example shown<br /> below). In addition, we evaluated over 50 glycopeptide MS2 spectra and found<br /> this error to be consistently made.

Resulting from these incorrect product ion assignments, the<br /> scores given in Byonic are artificially inflated. This error significantly<br /> impacts subsequent quality control filtering used by the authors especially<br /> score cutoffs. This error subsequently affects all figures in the manuscript as<br /> the number of identifications, and the quality of those identifications, is<br /> artificially high.

Another important step in annotating product ions to<br /> glycopeptide fragments is to adequately assign theoretical glycopeptide<br /> fragments to the observed product ions. As alluded to above, a total of 1084<br /> m/z values were observed for the spectrum in Figure 1A and this is artificially<br /> inflated due to the isotope distributions for each glycopeptide fragment. If we<br /> deisotope the spectrum to the best of our ability, we reduce the total number<br /> by 70% to 372 m/z values with a total of 75 annotations. While we acknowledge<br /> that MS2 deisotoping likely does not remove all non-monoisotopic peaks, this<br /> leaves over 80% of all observed monoisotopic peaks unannotated for the MS2<br /> spectrum in Figure 1A.

This significantly large number of unassigned product ions<br /> demonstrates that there are likely alternative fragmentation mechanisms<br /> occurring that have not been considered for spectrum annotation and/or scoring<br /> the given spectrum for its ability to describe the theoretical glycopeptide<br /> match. This is especially concerning because fragmentation events that have not<br /> been accounted for may be responsible for product ions annotated as simple<br /> b/c/y/z fragments due to their isomericity to these fragments.

Reviewer 3 gave a more than reasonable critique highlighting some of these issues and the subsequent impact on localization.This resulted in the authors removing any statements regarding localization in glycopeptides with multiple sites. Despite these efforts, we do not believe that the authors have taken appropriate action to mitigate the impact of<br /> erroneous product ion annotation (~20%) or significantly lacking assignment of<br /> glycopeptide product ions (~80% unannotated monoisotopic peaks in MS2 spectra)<br /> and therefore request a re-analysis of all product ion assignment in this<br /> manuscript with these critiques in mind.

On 2019-04-22 13:26:54, user Ranjan Nanda wrote:

Is so little sample was used for extraction or could be a typo error!<br /> Line 94: " Around 0.1 to 0.2 mg of frozen faecal sample was used for DNA extraction."

On 2019-04-22 07:24:22, user Francois wrote:

It lacks some supplementary tables.

On 2019-04-20 18:51:44, user Hanka wrote:

I'd just like to comment on the salary differences, since that received some publicity.

I think your analysis is misleading due to grouping by European region and sample being not representative. It seems to me that your results get skewed due to high participation of researchers from countries with good postdoc salaries (UK, Sweden, Germany). For example, in France, touching 40k a year is not realistic. postdoc salary in CNRS (the largest research body in France) is about 32k. (Which actually is not such a bad salary for France.) But only 23 French postdocs participated in the survey, as opposed to 178 British or 162 Swedish. I think, if you really want to analyze by region, you'd better estimate the number of postdocs in each country and use a weighted metric. However, I do think it would be better to break the results down by countries, include some kind of normalization by purchasing power and also focus a bit more on net income as that can be quite different based on tax progression. Finally, your sample from Eastern European region is very small, only 36 postdocs in total. It would seem tricky to draw many conclusions about that region from so few replies. Don't you think?

On 2019-04-19 21:58:25, user Charles Warden wrote:

I think your algorithm has a similar goal (and, to some extent, name) as derfinder

https://bioconductor.org/pa...

Perhaps a citation and comparison may be good to add?

On 2019-04-19 21:10:00, user Jared Ellefson wrote:

Might also want to consider citing this paper! :) https://www.ncbi.nlm.nih.go...

On 2019-04-19 20:47:05, user Stephanie Gogarten wrote:

It will be interesting to see how much this can speed up current WGS analysis pipelines that rely on VCF!

The reference to "Stilp et al." is incorrect, it should be listed as "Zheng et al.".

On 2019-04-19 11:29:56, user Eyal M. wrote:

This preprint is to be published in Cell Reports soon. The follow up of this study is published in Molecular Cell: https://www.cell.com/molecu...

On 2019-04-18 23:36:23, user Bhaskara Govinal wrote:

The function of AT1G16220 (AtPP2C6) is unknown in Arabidopsis. It is nice to see the characterization of tomato ortholog of it for plant immunity. Although this gene is named as PP2C6 in the genome-wide analysis published by Xue et al.,2008, many other reviews (Schweighofer et al., 2004; Fuchs et al., 2013) did not name this PP2C to show its function remains unknown. Further, name PP2C6 may lead to future confusion as there are other two PP2Cs in the same Clade already called as PP2C6-6 and PP2C6-7. How about author naming this PP2C based on its immunity related function or interaction with kinases.

On 2019-04-18 19:55:07, user UAB Bacteriology Journal Club wrote:

Very interesting paper! We (the University of Alabama at Birmingham Bacterial Pathogenesis and Physiology Journal Club) read this paper this week, and wanted to share our comments and suggestions, in the hopes that they will be helpful to you. We had a great discussion, and definitely wanted thank you for posting it!

Review of Limoli et al. “Interspecies signaling generated exploratory motility in Pseudomonas aeruginosa”

University of Alabama at Birmingham Bacterial Pathogenesis and Physiology Journal Club<br /> April 18, 2019

Summary:<br /> In this work, Limoli et al. show that both type 4 pili and flagella are involved in a significant change in P. aeruginosa motility in the presence of S. aureus, in which P. aeruginosa are able to surround and disperse S. aureus microcolonies. They also report that this motility effect is modulated by the S. aureus Agr quorum sensing system and that the presence of S. aureus affects cyclic AMP levels in P. aeruginosa. Finally, they confirm that interspecies signaling occurs with both laboratory and clinical isolates of both bacterial species.

This paper is fairly well-written, and explores an important and impactful topic. Understanding the molecular mechanisms of polymicrobial interactions is undeniably important. The microscopic movies illustrating changes in bacterial motility are powerful and compelling, and the authors do a good job of applying quantitative measures (e.g. Euclidean direction) to these otherwise qualitative observations. However, the second half of the manuscript is quite confusing, and the explorations of the role(s) of Agr and PilJ in Pseudomonas-Staphylococcus interactions lacked both depth and context. There were several instances of important conclusions that did not appear to be supported by the data presented, and in some cases the still images did not do a good job of representing the phenotypes observed in the movies (although we recognize that this is a significant challenge!).

Overall, although it has some substantial weaknesses, we feel that this paper presents fascinating and important observations, and we recommend that, for publication, the authors focus on expanding either the Agr or cAMP/PilJ studies, saving the other for future papers. We hope the following detailed comments will be helpful:

Major Comments:<br /> Figure 1 / Movies 1-3: Please explain (or at least cite!) why P. aeruginosa inhibits S. aureus growth. It is not clear what value the green and yellow “founder cell” markers add to Figure 1, or why the P. aeruginosa founder cell isn’t moving.

Figure 2 / Movie 4: The disassembly of S. aureus microcolonies is shown only in a single still image. This is an important result, and should certainly be included as part of a movie or other more in-depth data presentation. “Swift motion” is very confusing, and it is unclear what the authors mean precisely by this, since it doesn’t appear to have any relationship to speed of movement. The movie presented to illustrate this was difficult to find any movement at all in. We also wondered if there is any way to distinguish between “invasion” of an S. aureus microcolony by P. aeruginosa and “climbing” onto the microcolony (and therefore moving up out of the plane of focus), since S. aureus colonies are clearly 3-dimensional piles of cells. mKO is never defined (presumably it is a red fluorescent protein).

Figure 3 / Moves 5 – 8: The wild-type and ∆flgK still images appear to be from the wrong movies (these two in particular seem to be swapped). We found several instances throughout the paper of still images which did not appear to match with movies, and this was very troubling. Please check carefully to make sure that they are showing what’s intended. It would be very helpful to have still images and movies with matching fields of view. The timescales for movies 5 – 8 are very short, and make it difficult to assess the changes in the overall motility phenotypes. It would be very useful if the authors would comment on why the ∆pilA ∆flgK mutant is unable to inhibit S. aureus growth. The labeling of statistical significance in Figure 3B is unconventional, and inconsistent with the use of lowercase letters as labels in other figures. Please be consistent and clear about what comparisons are being made in all figures. Finally, the motions illustrated in Figure 3C are not very clear, and at least one of the labeled cells (single cell with red arrow) does not appear to be in motion at all.

Figure 4: The main concern with this figure was that neither twitching motility, sigB, nor Agr were introduced or explained in any detail. The rationale behind these experiments is not clear, and especially in the case of the very complex and well-studied Agr quorum sensing system, require much more information to understand the context and interpretation of these experiments. We were very curious as to why the authors did not directly test whether the Agr-dependent AIP signaling peptides are sensed by P. aeruginosa.

Figure 5 / Movie 9: The data in Figure 5A show a much more significant effect for Agr than does Movie 9, although upon repeated viewings, we think we understand the difference between P. aeruginosa behavior in Movie 4 and Movie 9, which appears to be primarily a disassembly of the P. aeruginosa raft even at very long distances from the S. aureus microcolony. Is this what the authors mean by “avoidance”? The sentences describing these results, especially “some cells seemed to actively avoid the S. aureus colony all together”, were not clear and did not seem to be supported by the data presented. Figure 5B certainly needs a wild-type control, and a better explanation of how to interpret 5C would be appreciated.

Figure 6: We were in general very confused by all of the results discussing PilJ and cAMP, and feel that the context, background, and interpretation of these experiments require much more explanation.

Minor Comments:<br /> 1) In the title, we would recommend “stimulates” rather than “generates”.

2) How has the polymicrobial nature of infections been clear since the 1600’s, when the germ theory of disease was established in the 1880s?

3) Typo in motility results section: “Through 1.5% agar”.

4) How close do P. aeruginosa need to be to S. aureus to experience a change in motility? Is this consistent with the known ability of small molecules / proteins to diffuse in agar?

5) There was some concern with the use of area to measure S. aureus replication, since the S. aureus colonies appeared to be generally taller than the rafts of P. aeruginosa, which we interpreted to indicate that they were mounding up on top of one another as they divide.

6) The twitching motility (under the agar) assay was unfamiliar to us, and we would have liked to see some experiments validating the relevance of this kind of motility to the single-cell microscopic observations the paper begins with

On 2019-04-18 18:06:07, user Paul Schanda wrote:

This is a really useful development. We have used FLYA ourselves and managed to assign a 12 x 39 kDa protein from solid-state NMR data in a fully automatic manner - the largest protein so far assigned in solids. About a year of manual work -- or a few hours with FLYA, leading to the identical result. (https://www.biorxiv.org/con...

Having this tool now - finally - for methyl assignments is very welcome and I am sure it will boost solution-NMR of large proteins further.

On 2019-04-18 17:57:11, user Paul Macklin wrote:

accepted!

On 2019-04-17 13:17:39, user Daniel Louis Kiss wrote:

The final peer-reviewed manuscript is available through FEBS Letter's free access program.<br /> https://twitter.com/FEBS_Le...

On 2019-03-25 17:11:09, user Daniel Louis Kiss wrote:

Also, the final paper is out: https://www.ncbi.nlm.nih.go...

On 2019-03-25 17:10:12, user Daniel Louis Kiss wrote:

Validation data for new primers used for PCR and qPCR in this paper plus replicate data for agarose gels and western blots are up on Figshare at the following link: <br /> https://figshare.com/articl...

On 2019-04-16 20:29:02, user Kasthuri wrote:

Please see the revised version of this article at:<br /> https://www.biorxiv.org/con...

On 2019-04-16 19:13:54, user weiya xue wrote:

how to reverse the chromosome order?

On 2019-04-16 18:20:04, user Charles Warden wrote:

Hi,

Thank you for sharing this preprint (and the code for analysis).

I hope I am not overlooking something, but I had a few questions:

1) I apologize, but I may be unclear about the experiment design. At first, I saw the “1,011 yeast isolates,” and guessed that you were doing an experiment for one generation for the 16 x 16 crosses (with 3-4 replicates each). However, I think I am supposed to be focusing on “13,950 individual recombinant haploid yeast segregants by crossing each parental strain to two different strains and collecting an average of 872 progeny per cross (Fig. 1, Supplementary Table 1).” Is there some sense of “generations” in the experiment, or are you looking for the effect of mutations in the parental strains in the original 16 x 16 crosses?

2) If you are looking for inherited mutations (from the parental strain), should the minimum MAF for a given variant be closer to 6% (1/16) or 3% (1/32, for haploid segregates from diploid parents, if I remember the yeast biology right)? If so, I’m sorry, but I’m a little confused about the <1% (inherited / heritable) variants.

3) If you have a novel mutation causing a more clear phenotype (such as an acquired somatic mutation that is selected for as resistance to your treatments?), are you testing additional crosses/replicates with that isolate (to see if you can reproduce the phenotype in other crosses/segregants that retain that mutation)? Or, does that have something to do with 100s of segregants per cross (for example, testing the phenotype for a multiple isolates at X weeks, 2X weeks, 3X weeks, etc.)?

4) Is there an accession number for deposited data?

5) Also, it probably doesn’t need to be said for the yeast readers, but I was admittedly almost confused by chrX in Figure 4 (but I eventually figured out that was the 10th chromosome).

Thank you very much for taking time to review these questions, particularly from somebody who doesn’t perform an experiment like this on a regular basis.

Best of luck with your lab’s projects!

Sincerely,<br /> Charles

On 2019-04-16 14:35:28, user hugo wrote:

Hello, the github link is not working. Maybe is this repo? https://www.biorxiv.org/con...

Best regards.

On 2019-04-16 06:47:10, user Veranja Liyanapathirana wrote:

Good piece of work but what about viruses that are equally imporimport to identify? Would your human depletion pipeline remove them? Is there a way to assess the quality of the samples to screen proper sputum samples as opposed to salivary ones that we often get to diagnostic labs inbuilt to this?

On 2019-04-15 20:11:49, user Christos Gournas wrote:

Very interesting work! And very consistent with our recent work on the Can1 Arg transporter of S. cerevisiae. We have recently published two stories about the substrate-induced endocytosis of Can1 (see below). In this work, we provide evidence that a substitution of E184 in TM3 by Gln (so E184Q) should block the transporter in an IF conformation. As the presence of Gin should mimic permanent protonation of E184, the results presented herein seem to be at the same direction with ours! I'm really excited !<br /> Gournas, C., Saliba, E., Krammer, E.-M., Barthelemy, C., Prévost, M., and André, B. (2017). Transition of yeast Can1 transporter to the inward-facing state unveils an α-arrestin target sequence promoting its ubiquitylation and endocytosis. Mol. Biol. Cell 28, mbc.E17-02-0104.

Gournas, C., Gkionis, S., Carquin, M., Twyffels, L., Tyteca, D., and André, B. (2018). Conformation-dependent partitioning of yeast nutrient transporters into starvation-protective membrane domains. Proc. Natl. Acad. Sci. 115, E3145–E3154.

On 2019-04-15 18:58:02, user Michael McLaren wrote:

This looks like a very valuable resource for the community! Can you clarify the status of the data availability? The study identifier given in the manuscript does not seem to be currently available (https://www.ebi.ac.uk/ena/d.... Also, which of the non-human data will be available in the ENA and which is to be made available on request? Thanks!

On 2019-04-15 09:40:53, user Albert Bondt wrote:

Hi,<br /> Nice systematic approach!<br /> I hope you don't mind if I ask a few questions though...

You say you avoid enrichment to avoid introducing a bias. However, you do perform reduction/alkylation, precipitation, and drying/reconstitution. These steps can also introduce a bias. While for IgG glycopeptides this is not required (check out the many papers by Manfred Wuhrer on IgG glycosylation, since 5 years or so also using a NanoBooster). So did you check whether these sample prep steps do not introduce a bias, or are required to avoid bias?

Another issue is for Figure 3. While the peptide backbone of IgG2 and IgG3 Variant are identical, the glycosylation profile is different. Or did you use purified single subclasses (not clear from Materials and Methods)? In case of the latter, did you check the purity? IgG3 purification remains a major challenge, it is often contaminated with other subclasses.

On 2019-04-14 11:24:46, user 宮田真人 wrote:

Thank you for your interests on quick freeze, deep etch EM.<br /> We believe this method is really useful for PG studies.

We have been studying about motility mechanisms of class Mollicutes.<br /> Recently, we got interests on the role of PG on survival and evolution, because class Mollicutes quit it by some reason.

We hope we can do some contributions to PG field.<br /> Any comments are welcome. <br /> Makoto MIYATA

On 2019-04-14 08:05:38, user Arpan Parichha wrote:

I was thinking of using the SABER technique in our system and wanted use its multiplexing feature for our purpose. I am unable to use the oligominer pipeline. Can anybody tell me how to design the PER primers?

On 2019-04-13 17:11:43, user Andrei I Ivanov wrote:

We have published a paper a few years ago that shows the roles of anillin in regulating apical junctions and the perijectonal actomyosin in human epithelila cells https://www.ncbi.nlm.nih.go...<br /> I am surprised that the authors did not refer to our study.

On 2019-04-13 14:05:34, user Javier Forment wrote:

The PDF of Data Supplement at Supplementary Material indicates that there exists a "Supplemental dataset 2: NbC gene-models database gff3 annotation". I cannot find that gff3 file anywhere. Could you please tell me what I'm doing wrong?

On 2019-04-13 11:25:00, user Jingpeng Wu wrote:

it would be better to explain how do you achieve the performance in the abstract. It took me some time to find it out, and it is still not quite clear for me.