Dear %TITLE% Ventre,

Thank you very much for submitting your manuscript "One model fits all: combining inference and simulation of gene regulatory networks" 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.

You will see that the reviewers have appreciated the topic and the writing of the manuscript. However, some of the reviewers ask for more details on specific computational details, and suggest some clarifications in the explanation of the methods.

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,

Carl Herrmann, Ph.D.

Academic Editor

PLOS Computational Biology

Jason Haugh

Section 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:

Reviewer's Responses to Questions

Comments to the Authors:

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

Reviewer #1: The authors present a computational technique for gene regulatory network (GRN)-based simulation of single-cell transcriptomics data sets, with the underlying GRN being reconstructed from a real data set. This functionality is not present in existing GRN-based simulation tools.

The tool called HARISSA, published previously, is a very carefully designed mechanism-based simulator of single-cell expression dynamics under the influence of a GRN. The tool called CARDAMON, also previously published, is capable of inferring the GRN underlying a single cell transcriptomic data set, under the same mechanistic model used by HARISSA.

Evaluation of GRN inference is performed first on synthetic data generated by HARISSA. In these evaluations, the CARDAMOM algorithm previously published by the authors performs best among several appropriate competing methods. This demonstrates that the CARDAMOM method is indeed able to reconstruct GRNs when the synthetic data arise from its underlying model, as expected.

The authors then apply their methods to partially reconstructing the GRN for Retinoic Acid-induced differentiation of mouse ESCs and validate it based on ChIP data, again noting improvements over appropriate baseline (existing) methods. This is a useful exercise for a future user of the tool, even though the actual performance metrics (AUPRC, AUROC ~ 0.6 compared to ~0.5 random baseline) are not stellar.

The last section is a very interesting and exciting aspect of the work, where networks inferred from real data are then used to simulate synthetic data which is shown to match the real data. This is a clear advance in the field.

Overall, this study is a rigorous and innovative next step at the forefront of scRNA-seq data modeling and simulation, and to some extent of GRN reconstruction as well. I wholeheartedly recommend publication of the work.

Reviewer #2: In the manuscript the authors combine two computational tools for the simulation and inference of GRNs and test them in multiple scenarios including in silico and in vitro data. The results are analyzed with rigorous statistical tools and are well organized.

I appreciate the effort that the authors put on the analysis specially comparing with other existent tools, I think that this is crucial for good research on this topic were new algorithms are appearing continuously. While I am familiar with the simulation and inference of GRN, I am not familiar with the state of the art of libraries applied to scRNAseq data and therefore it is hard for me to identify the novelty of the manuscript. Both tools used in this manuscript HARISSA and CARDAMOM have been developed in the past. I would have appreciated to have a more explicit introduction to the technicalities of the different tools used and a more clear identification of the differences to previous publications on these two tools.

Reviewer #3: This well-written manuscript presents a compelling new approach to predict gene regulatory networks (GRNs) by combining two previously-published methods HARISSA and CARDAMOM. The key component of HARISSA that is of use here is a GRN simulation algorithm based on a "multi-agent" generalization of a standard two-state stochastic, mechanistic model of gene expression. This model captures bursting transcriptional dynamics. CARDAMOM is a scalable algorithm for inference of the parameters of this model, and hence, the underlying GRN itself. In combination, these methods permit both inference of a GRN and its simulation.

The analysis is very thorough. The choice of algorithms for comparison is very good. The authors first focus on small, synthetic, networks with well-understood dynamics. HARISSA and CARDAMOM clearly outperform other algorithms and when they do not, it is clear that even a simple correlation-based algorithm can perform well. The decision to use precision-recall curves here is cogent. I liked the care given to SCRIBE, the scalability analysis for trees, and the thorough analysis of the limit of performance with respect to the number of cells and timepoints.

The analysis of the experimental data is also thorough and well-considered. The performance (AUPRC) of HARISSA and CARDAMOM drops to around 0.6 but is still larger than the other algorithms. The highlight of this manuscript is the simulation of the inferred network and the comparison of the resulting data to the original measurements. This analysis, which is rarely seen in GRN inference papers (mainly because it is simply not possible), will make this manuscript stand out when published.

The interpretation of results is very careful. The authors

I have a few suggestions for further strengthening the manuscript.

1. The authors should clarify whether they used every pair of genes not present in the ground truth as a negative example to compute the number of false positive and true negatives. In this context, I am puzzled as to how the random classifier can have a precision of 0.47 in Figure 4B. This high value suggests a ground truth network where nearly half the possible edges are present.

2. The authors should include a supplement that studies the network inferred by GENIE3 (or SINCERITIES) in a manner similar to page 11 and Figure 5.

3. In Figure 7A, the p-values should be corrected for testing multiple hypotheses. I am not sure if they are. The authors could reduce the number of tests by performing one K-S test for each gene considering its expression over all time points. It will also be instructive to see such a plot for the naive model without interactions.

4. On page 3, lines 79-80, the authors state that "existing GRN-based simulation tools, which are generally based on more phenomenological than mechanistic models". It may be that my understanding of "phenomenological" and "mechanistic" is different, but isn't BoolODE based on simulating a Boolean (mechanistic) model rather than a phenomenological one? In Discussion, the authors state that SERGIO uses a mechanistic model.

MINOR COMMENTS

page 18, line 447: "by going up the arrow of time" may be better phrased as "by going back in time".

page 21, line 558: Change "bursts" to "burst".

page 21, line 561: The meaning of the phrase "a synthetic noise well adapted" is unclear. Perhaps this sentence should be rephrased.

Several citations are missing page numbers.

Reviewer #4: Review in attachment

**********

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: No: The software for both HARISSA and CARDAMOM are available but I do not see the network files for the toy networks in Figure 1, the simulated data for the corresponding analyses, or the ground truth networks for Figure 4. The authors should provide code to generate all the results in their manuscript, even if in the form of Jupyter notebooks.

Reviewer #4: 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

Reviewer #4: 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

References:

Review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript.

If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Attachments

Attachment

Submitted filename: Report.pdf

Dear %TITLE% Ventre,

We are pleased to inform you that your manuscript 'One model fits all: combining inference and simulation of gene regulatory networks' 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,

Carl Herrmann, Ph.D.

Academic Editor

PLOS Computational Biology

Jason Haugh

Section 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 revision is very good. The authors have addressed my concerns well.

Reviewer #4: The authors have clearly answered the remarks and questions and have amended the manuscript which has clarified the text

**********

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 #3: Yes

Reviewer #4: 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

Reviewer #4: No