Dear Dr. Youngblut,

Thank you very much for submitting your manuscript "ResMiCo: increasing the quality of metagenome-assembled genomes with deep learning" for consideration at PLOS Computational Biology.

As with all papers reviewed by the journal, your manuscript was reviewed by members of the editorial board and by several independent reviewers. In light of the reviews (below this email), we would like to invite the resubmission of a significantly-revised version that takes into account the reviewers' comments.

The reviewers all agree that the methods are an improvement over the state-of-the-art and that the tool could potentially be widely used, but raise several important points related to both unclear presentation and whether some of the conclusions are sufficiently supported. Of particular interest to us are the concerns related to the evaluation on real data: the suggestion of reviewer #3 to test ResMiCo on data with a known standard seems very reasonable; similarly, the comment of reviewer #1 that only human gut samples are used should be addressed (ideally by including additional benchmarks).

We cannot make any decision about publication until we have seen the revised manuscript and your response to the reviewers' comments. Your revised manuscript is also likely to be sent to reviewers for further evaluation.

When you are ready to resubmit, please upload the following:

[1] A letter containing a detailed list of your responses to the review comments and a description of the changes you have made in the manuscript. Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

[2] Two versions of the revised manuscript: one with either highlights or tracked changes denoting where the text has been changed; the other a clean version (uploaded as the manuscript file).

Important additional instructions are given below your reviewer comments.

Please prepare and submit your revised manuscript within 60 days. If you anticipate any delay, please let us know the expected resubmission date by replying to this email. Please note that revised manuscripts received after the 60-day due date may require evaluation and peer review similar to newly submitted manuscripts.

Thank you again for your submission. We hope that our editorial process has been constructive so far, and we welcome your feedback at any time. Please don't hesitate to contact us if you have any questions or comments.

Sincerely,

Luis Pedro Coelho

Academic Editor

PLOS Computational Biology

William Noble

Section Editor

PLOS Computational Biology

***********************

The reviewers all agree that the methods are an improvement over the state-of-the-art and that the tool could potentially be widely used, but raise several important points related to both unclear presentation and whether some of the conclusions are sufficiently supported. Of particular interest to us are the concerns related to the evaluation on real data: the suggestion of reviewer #3 to test ResMiCo on data with a known standard seems very reasonable; similarly, the comment of reviewer #1 that only human gut samples are used should be addressed (ideally by including additional benchmarks).

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: The review is uploaded as an attachment.

Reviewer #2: This paper addresses the major challenge of evaluating contigs misassembly from the metagenomic reads without reference requirements. The proposed system, ResMico, utilizes a novel deep convolutional neural network with skip connections between non-adjacent layers. The authors have performed extensive evaluations through simulated datasets representing variously plausible microbial communities and sequencing parameters utilizing reference bacterial and archaeal genomes from the Genome Taxonomy Database. They showed the superior performance (accuracy and robustness) of ResMico compared to the existing methods, including DeepMAsDB, the tool previously developed by the authors’ group. They further showcased ResMiCo’s ability to detect an average of 4.7% error rate of contigs/metagenome in a large collection of gut metagenomes. The paper presents a tool with significant utility in metagenomics research. It is generally well-written; however, some technical aspects could be better explained.

1) The Residual block structure was used in ResMiCo. However, more detailed descriptions are needed to clarify the rationale for using it, and how the number of RGs was determined. Please also indicate in the Fig.2 caption that 14 represents the number of the selected features in the input to the neural network.

2) P9: Clarify why MetaSPAdes (but not MEGAHIT) was used for assembling the published, real-world metagenomes.

3) P12: the number of contigs used in the feature importance evaluation seems relatively small: 200 contigs randomly selected from the correct and incorrect. There is a concern if the size is sufficiently large for a good assessment of the importance. Could you discuss further on this?

4) P15: Using UMAP to examine how data were represented by ResMiCo is a good idea. However, as we know, UMAP only allows a visual inspection. I wonder if clustering the misassembled contigs would generate additional insights on the type of misassemblies in the data and if ResMiCo makes more mistakes in classification for any kind of misassembles represented by some cluster.

Reviewer #3: In this paper, the authors present ResMiCo, a deep convolutional neural network that is trained to identify misassembled contigs in a reference-free manner, from metagenome assemblies. While metagenome assemblies are rapidly accumulating, the quality of the assemblies can still be much improved in many cases. In this respect, it is critical to identify and correct misassembled contigs from existing metagenomes. As per results, ResMiCo outperforms prior methods *greatly*. While this is very positive, we have remained not yet fully convinced in terms of the experiments that support the idea of superiority. In the following some major comments.

* In lines 285-286, ”This filtering resulted in a reduction of the true error rate from 4% to 1% while keeping the contiguity metrics virtually unmodified.” How exactly do you estimate the error rate?

* In Metaquast, how does Genome fraction change after filtering out misassembled contigs?

* It would be clearly favorable to evaluate the method on real data that have a standard reference. Suggestions are: the Bmock12 data set in Volkan Sevim’s work treating shotgun metagenome data reflecting a mock community, sequenced using Oxford Nanopore, PacBio and Illumina technologies. Another dataset is the NWCs (natural whey culture) data set in Vincent Somerville’s “Long-read based de novo assembly of low-complexity metagenome samples results in finished genomes and reveals insights into strain diversity and an active phage system.” In both works, short reads and standard references are provided, where the latter was generated using hybrid assembly methods.

* Please provide Genome Fraction as an evaluation category generated by Metaquast. Since metagenomes often contain various strains referring to identical species, it is important to have a look at how Genome Fraction is affected. For example, loosing contigs from particular strains -- indicated by a drop in Genome Fraction -- would imply undesired biases.

**********

Have the authors made all data and (if applicable) computational code underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data and code underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data and code should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data or code —e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

Figure Files:

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

Data Requirements:

Please note that, as a condition of publication, PLOS' data policy requires that you make available all data used to draw the conclusions outlined in your manuscript. Data must be deposited in an appropriate repository, included within the body of the manuscript, or uploaded as supporting information. This includes all numerical values that were used to generate graphs, histograms etc.. For an example in PLOS Biology see here: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001908#s5.

Reproducibility:

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Attachments

Attachment

Submitted filename: renamed_4c385.pdf

Dear Dr. Youngblut,

Thank you very much for submitting your manuscript "ResMiCo: increasing the quality of metagenome-assembled genomes with deep learning" for consideration at PLOS Computational Biology.

As with all papers reviewed by the journal, your manuscript was reviewed by members of the editorial board and by several independent reviewers. In light of the reviews (below this email), we would like to invite the resubmission of a significantly-revised version that takes into account the reviewers' comments.

We ask that the authors address the remaining concerns of reviewer #3.

We cannot make any decision about publication until we have seen the revised manuscript and your response to the reviewers' comments. Your revised manuscript is also likely to be sent to reviewers for further evaluation.

When you are ready to resubmit, please upload the following:

[1] A letter containing a detailed list of your responses to the review comments and a description of the changes you have made in the manuscript. Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

[2] Two versions of the revised manuscript: one with either highlights or tracked changes denoting where the text has been changed; the other a clean version (uploaded as the manuscript file).

Important additional instructions are given below your reviewer comments.

Please prepare and submit your revised manuscript within 60 days. If you anticipate any delay, please let us know the expected resubmission date by replying to this email. Please note that revised manuscripts received after the 60-day due date may require evaluation and peer review similar to newly submitted manuscripts.

Thank you again for your submission. We hope that our editorial process has been constructive so far, and we welcome your feedback at any time. Please don't hesitate to contact us if you have any questions or comments.

Sincerely,

Luis Pedro Coelho

Academic Editor

PLOS Computational Biology

William Noble

Section Editor

PLOS Computational Biology

***********************

We ask that the authors address the remaining concerns of reviewer #3.

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: Thanks to the authors for addressing all of my concerns and comments from the first review in the revised manuscript. I have no remaining questions and wish the tool will be well maintained and developed continuously in the future.

Reviewer #2: All my comments and suggestions have been satisfactorily addressed.

Reviewer #3: I disagree with the authors on not running the method on mock communities. While these mock communities are not *real-world* data, they are *real* data, and provide -- as widely accepted by the community -- a basic test that a method does not perform in an erratic manner when being confronted with real data. (thanks for listing the communities, but I am aware of everything you mention in your response anyway)

So, please run your method on mock communities, or convince me otherwise that it does not behave in mistaken ways when confronted with *real* data. (there was no mentioning in the response that you had done this; and I sometimes had the feeling that there was some confusion about the difference between *real-world* and *real* data).

Thank you very much.

**********

Have the authors made all data and (if applicable) computational code underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data and code underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data and code should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data or code —e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

Figure Files:

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

Data Requirements:

Please note that, as a condition of publication, PLOS' data policy requires that you make available all data used to draw the conclusions outlined in your manuscript. Data must be deposited in an appropriate repository, included within the body of the manuscript, or uploaded as supporting information. This includes all numerical values that were used to generate graphs, histograms etc.. For an example in PLOS Biology see here: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001908#s5.

Reproducibility:

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Dear Dr. Youngblut,

We are pleased to inform you that your manuscript 'ResMiCo: increasing the quality of metagenome-assembled genomes with deep learning' has been provisionally accepted for publication in PLOS Computational Biology.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. A member of our team will be in touch with a set of requests.

Please note that your manuscript will not be scheduled for publication until you have made the required changes, so a swift response is appreciated.

IMPORTANT: The editorial review process is now complete. PLOS will only permit corrections to spelling, formatting or significant scientific errors from this point onwards. Requests for major changes, or any which affect the scientific understanding of your work, will cause delays to the publication date of your manuscript.

Should you, your institution's press office or the journal office choose to press release your paper, you will automatically be opted out of early publication. We ask that you notify us now if you or your institution is planning to press release the article. All press must be co-ordinated with PLOS.

Thank you again for supporting Open Access publishing; we are looking forward to publishing your work in PLOS Computational Biology. 

Best regards,

Luis Pedro Coelho

Academic Editor

PLOS Computational Biology

William Noble

Section Editor

PLOS Computational Biology

***********************************************************