Dear Dr Finnie,

Thank you very much for submitting your manuscript "Backtracking: Improved methods for identifying the source of a deliberate release of Bacillus anthracis from the temporal and spatial distribution of cases" 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.

First, apologies for the long delay in reaching this decision. We had a very hard time securing a second reviewer. As you will see, the reviews are mixed. In your response and revision, please pay particular attention to Reviewer 2's comment about providing more thorough background (literature review) and comparison to existing methods, including formal metrics for the model fit, as well as a better description of the methods in the main text. Please also make a better effort to adhere to our journal's code-sharing policy, as the reason for not sharing the code seems inadequate.

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,

Virginia E. Pitzer, Sc.D.

Section Editor

PLOS Computational Biology

Virginia Pitzer

Section Editor

PLOS Computational Biology

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

First, apologies for the long delay in reaching this decision. We had a very hard time securing a second reviewer. As you will see, the reviews are mixed. In your response and revision, please pay particular attention to Reviewer 2's comment about providing more thorough background (literature review) and comparison to existing methods, including formal metrics for the model fit, as well as a better description of the methods in the main text. Please also make a better effort to adhere to our journal's code-sharing policy, as the reason for not sharing the code seems inadequate.

Reviewer's Responses to Questions

Comments to the Authors:

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

Reviewer #1: This article describes an alternative model calibration method than normally used in literature. As such it is well considered, timely and of interest.

I have the following comments prior to publication:

1) Line 22 - 24, 'two planar directions' reads oddly and planar later, perhaps 'across 2 dimensional plan is better'. Though the briggs model I think allows height as it is a Gaussian puff model with empirical diffusion terms.

2) Line 30 'realistic' may be too strong, it is certainly credible or plausible but the simplicity of Briggs given turbulent flow and population variation, with a lack of observational data of such an event means the simulation proposed is unvalidated. This does not undermine approach or results but may inflate readers expectations.

3) In intro the citation for within host model is [7] but in methods it becomes [8] can authors clarify.

4) Line 71 - reads oddly. "hospitalisations or deaths are recorded after the censor date, even if we know that" i think should be "hospitalisations or deaths are used after the censor date, even if we simulate that"

5) Line 75 - typo, no need for 'of'

6) Line 135 - " guaranteed not" is too strong, I think should be "not guaranteed" was intended, unless the authors have a citation for a proof.

7) Figures 4, 5, 6 - in cation can the authors explain the symbols and lines.

8) Line 292 - "the preparation of synthetic training data," please add something like "and validity of simulation model" as method would be prone to structural misspecification.

9) Line 311-12 - I agree but for grid search and MCMC the range of search/prior distribution definition may be considered given the structural assumptions in model to limit the range and chance of missing global maxima. This is going to potentially bias the results if the model is misspecified of course. Given the spatial pattern in Briggs is diffusion dominated after some distance most information is in the close cases - which if release was far from population centre may mean there is limited info in the data or likelihood may infer as may be equivalent to a far release. So work effort may be on the structure of the underlying model too.

Reviewer #2: The authors set out to improve upon reverse-epidemiological methods using a recurrent convolutional neural network approach. My comments are as follows.

1. This paper has no literature review. As a result, it is completely impossible to assess whether or not what the authors are contributing is a valuable contribution. For example, I did a quick google scholar search with the following search terms: “neural network” and “outbreak source location”, and saw a wide variety of papers. The geographic profiling literature has no doubt made contributions worth discussing as well. Additionally, the authors dedicate considerable time to discussing limitations to MCMC-based procedures, but don’t review them meaningfully. There is an entire body of statistical literature that addresses the same limitations to metropolis-hastings style algorithms that the authors note, but seemingly fail to incorporate in their work. Basically, the authors are trying to argue that they’ve developed a novel tool that out-performs existing methods. However, by not providing a meaningful literature review, they cannot convince this reviewer that their contribution is novel.

2. In the section introducing forward models (and the Briggs model, specifically), the authors need to introduce this better in the main text instead of the appendix. I do not feel it is sufficient for future readers of this paper to simply say that they use the Briggs’ model and then point to the appendix. I think the first page of the supplement would be better suited in the main text.

3. The RCNN is poorly described. For example, what’s an LSTM layer? It’s never described.

4. Results table 1. If these are R^2 values, why are some of them negative?

5. The entire results section is written as captions for figures and tables. Also, what are the criteria for a method being “good” or not having “a great deal of accuracy”?

6. I simply don’t understand the notion that releasing the code to this model is too much of security threat. In part this is because the authors have failed to convince me that they’ve made a novel contribution (see point 1 above). Do the authors think that if people who are prone to releasing anthrax in the UK had the code that detailed this modeling approach, they would alter their plans in any way? Not being an expert in this area, I assume their concerns are more focused on logistics than machine learning.

**********

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: No: They caveat this in submission. This is for editiorial team to judge on precedent rather than reveiewer.

Reviewer #2: No: No. This is a simulation study though.

**********

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

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 Finnie,

Thank you very much for submitting your manuscript "Backtracking: Improved methods for identifying the source of a deliberate release of Bacillus anthracis from the temporal and spatial distribution of cases" 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.

While the reviewer and editor find the revisions to be responsive to the issues raised in the first round of reviews, we still have concerns over the code availability, which we take very seriously at PLOS Computational Biology. After discussion with the Editors-in-Chief, we feel that providing no means of accessing the code is unacceptable. While we understand that the code cannot be made publicly available due to security concerns, for purposes of reproducibility and for purposes of utility, some mechanism of accessing the code should be provided, even if that requires a high bar such as approval/clearance by the UK government, NDAs/code control with the agency, etc. Alternatively, please consider whether it would be possible to provide a redacted and/or simplified version of the code based on simulated data that would not pose a security concern.

To address these concerns, please add a "Data and Code Availability" section at the end of the Methods that describes the reasons why the data/code cannot be shared publicly and includes all necessary contact information where an interested reader would need to apply in order to obtain the code.

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,

Virginia E. Pitzer, Sc.D.

Section Editor

PLOS Computational Biology

Virginia Pitzer

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:

While the reviewer and editor find the revisions to be responsive to the issues raised in the first round of reviews, we still have concerns over the code availability, which we take very seriously at PLOS Computational Biology. After discussion with the Editors-in-Chief, we feel that providing no means of accessing the code is unacceptable. While we understand that the code cannot be made publicly available due to security concerns, for purposes of reproducibility and for purposes of utility, some mechanism of accessing the code should be provided, even if that requires a high bar such as approval/clearance by the UK government, NDAs/code control with the agency, etc. Alternatively, please consider whether it would be possible to provide a redacted and/or simplified version of the code based on simulated data that would not pose a security concern.

To address these concerns, please add a "Data and Code Availability" section at the end of the Methods that describes the reasons why the data/code cannot be shared publicly and includes all necessary contact information where an interested reader would need to apply in order to obtain the code.

Reviewer's Responses to Questions

Comments to the Authors:

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

Reviewer #1: The authors have attended to all my comments and questions sufficiently.

**********

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

**********

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

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.

Dear Dr Finnie,

We are pleased to inform you that your manuscript 'Backtracking: Improved methods for identifying the source of a deliberate release of Bacillus anthracis from the temporal and spatial distribution of cases' 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,

Virginia E. Pitzer, Sc.D.

Section Editor

PLOS Computational Biology

Virginia Pitzer

Section Editor

PLOS Computational Biology

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