Dear Bikard,

Thank you very much for submitting your manuscript "Generating functional protein variants with variational autoencoders" 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. The reviewers appreciated the attention to an important topic. Based on the reviews, we are likely to accept this manuscript for publication, providing that you modify the manuscript according to the review recommendations.

 In particular there is a need to address remaining concerns around controlling the solubility level and  baseline comparisons.

Please prepare and submit your revised manuscript within 30 days. If you anticipate any delay, please let us know the expected resubmission date by replying to this email. 

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

[1] A letter containing a detailed list of your responses to all 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.

Thank you again for your submission to our journal. 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,

Christine A. Orengo

Associate Editor

PLOS Computational Biology

Arne Elofsson

Deputy Editor

PLOS Computational Biology

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

A link appears below if there are any accompanying review attachments. If you believe any reviews to be missing, please contact ploscompbiol@plos.org immediately:

[LINK]

Reviewer's Responses to Questions

Comments to the Authors:

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

Reviewer #1: The authors addressed reviewers’ comments thoroughly. The revised version is improved in quality. I have no further suggestions to make.

Reviewer #2: The paper presents a study of generating sequences for luciferase domains based on training VAEs from a set of 70,000 protein sequences similar to luciferase sequences. Some sequences from each model are realised and their activity measured.

The generated sequences mostly do luminesce, though none as strongly as WT. [It would be good to know about more controls than just the one -are any of the 70k sequences better than the measured WT?)

The need to do this (given there are 70,000 examples) is not strongly motivated in itself, but the paper did go on to show conditional versions where solubility of the generated sequences is controlled, while retaining function, though it’s not shown whether this is better or worse than just directly modifying the WT to adjust solubility in ways that are well-known (especially given the extensive MSA - cf PROSS).

I find that the authors in both the paper and the response to previous reviewers’ comments do seem to consider “edit distance” a bit naively. Yes, 5 random mutations [uniform in position and across AA type] are very likely to destroy function, but it would not be hard to choose sensible mutations that are unlikely to destroy function, especially given such an extensive MSA.

I agree with the previous reviewers (and the authors) that there is a lot of work in this direction, and this is very limited by needing such a large dataset of similar proteins whose function we're just recapitulating. I do nevertheless find it worthy enough for publication. I find the contrast between the “MSA” and AR versions of the model interesting.

Reviewer #3: Overall, the authors adequately addressed the issues raised in the initial review and present a reasonable plan for a re-submission. There are, however, some remaining points that should be addressed:

To demonstrate the value of conditioning on solubility, and support the claim that the conditional model can be used to “control the solubility level while introducing relatively few mutations”, the authors show a plot which compares the predicted solubility of sequences generated by the conditional model and sequences in the training set. To better support for these claims, the authors should also include the predicted solubility of sequences generated by the model without conditioning on solubility. This would more directly address the value of using a conditional model.

To address a concern (raised by multiple reviewers) that the model should be more extensively compared to baselines, the authors present some additional analysis, where they show that the model ELBOs show a higher correlation with protein luminescence than BLOSUM or HMM scores. The results shown provide compelling evidence that the MSA VAE can generate novel variants that maintain luciferase activity. The HMM does, however, reasonably predict the luminescence of AR-VAE generated sequences, suggesting that one could potentially generate functional proteins by sampling from the HMM.

A final, minor, point is that the authors indicate that the sequences generated using the conditional model are more similar to the wild type sequence than sequences generated by the VAE without conditioning on solubility. Aggregate statistics for their luciferase assays are nonetheless reported in the abstract. This is misleading and is an overestimate of the performance of the non-conditional model. The number of functional variants that are far away from training sequences is less than the presented amount.

**********

Have all data underlying the figures and results presented in the manuscript been provided?

Large-scale datasets should be made available via a public repository as described in the PLOS Computational Biology data availability policy, and numerical data that underlies graphs or summary statistics should be provided in spreadsheet form as supporting information.

Reviewer #1: Yes

Reviewer #2: No: While code and network weights are provided, with a license, I did not find a pointer to the associated data.

I would expect to see the 3000 sequences generated from each model, the predicted solubiliities, as well as actual luminescence and solubility for each of the measured proteins.

Reviewer #3: None

**********

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, PLOS recommends that you deposit 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. For instructions see http://journals.plos.org/ploscompbiol/s/submission-guidelines#loc-materials-and-methods

Dear Dr Bikard,

We are pleased to inform you that your manuscript 'Generating functional protein variants with variational autoencoders' 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,

Christine A. Orengo

Associate Editor

PLOS Computational Biology

Arne Elofsson

Deputy Editor

PLOS Computational Biology

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

Reviewer's Responses to Questions

Comments to the Authors:

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

Reviewer #3: The authors addressed all my comments.

**********

Have all data underlying the figures and results presented in the manuscript been provided?

Large-scale datasets should be made available via a public repository as described in the PLOS Computational Biology data availability policy, and numerical data that underlies graphs or summary statistics should be provided in spreadsheet form as supporting information.

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 #3: No