Dear Ms Tammi,

Thank you very much for submitting your manuscript "Accurate multi-population imputation of MICA, MICB, HLA-E, HLA-F and HLA-G alleles from genome SNP data" 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.

The manuscript has been seen by two expert reviewers, and both find the work very valuable for the community. The reviewers suggest some improvements that I would encourage the authors to follow. Both reviewers address to point to which extent this imputation approach can be used on data from SNP arrays rather than WGS/WES data, and it would be important to see at least a discussion and clarification of this. If the authors think that their approach would also work with such less dense SNP data, they are welcome to provide additional test as suggested by reviewer 2.

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,

Tobias Lenz

Guest Editor

PLOS Computational Biology

Mark Alber

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:

The manuscript has been seen by two expert reviewers, and both find the work very valuable for the community. The reviewers suggest some improvements that I would encourage the authors to follow. Both reviewers address to point to which extent this imputation approach can be used on data from SNP arrays rather than WGS/WES data, and it would be important to see at least a discussion and clarification of this. If the authors think that their approach would also work with such less dense SNP data, they are welcome to provide additional test as suggested by reviewer 2.

Reviewer's Responses to Questions

Comments to the Authors:

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

Reviewer #1: In this work, Silja Tammi et al. perform a thorough investigation and development of an imputation pipeline for MICA, MICB, HLA-E, HLA-1 F and HLA-G (including UTR polymorphism). The resources used are multiethnic and include the 1000 genomes and Finn Gen. I have only a few comments:

Having an idea of the allele frequency of the predicted alleles is important as imputation can be very accurate if polymorphisms are rare. This is indirectly presented in the figures with the counts, which I particularly appreciated, together with the availability of the data.

They could have further validated their algorithm using the UKB data

There are 2 limitations of this work, one clearly stated the other less so.

Then first limitation pertains to the fact a lot of incomplete coverage sequence had to be eliminated. This could create a bias, and possibly having an effect on rare alleles, but this is acknowledged.

The second limitation has to do with the scope of the study. The title and the abstract, unless read carefully, could be misinterpreted as if the authors were establishing a high accuracy imputation method for these gene alleles usable with whole genome association genotyping arrays. This is not the case, it is an imputation method to be used for exome or whole genome sequencing reads; this also explains the high accuracy as it uses dense SNP within the gene (+/- a few kb). In theory the whole coding information needed to define each allele is there except phasing in some cases where long reads could be needed.

It does not change the fact this is useful especially now that WGS and WES is mainstream and that questionable results such as in Commun Biol. 2023; 6: 1113 are starting to be published (the paper reports high frequencies of unknown alleles in the UKB). I would suggest that the title "Accurate multi-population imputation of HLA-E, HLA-F, HLA-G, MICA and MICB alleles from genome SNP data" be changed into "Accurate multi-population imputation of HLA-E, HLA-F, HLA-G, MICA and MICB alleles from genome SNP data derived from exome or whole genome sequencing".

Also, in the abstract, the sentence "To facilitate studies of the nonclassical and non-HLA genes in large patient and biobank cohorts, we trained imputation models for MICA, MICB, HLA-E, HLA-F and HLA-G alleles on genome SNP array data" could be changed into "To facilitate studies of the nonclassical and non-HLA genes in large patient and biobank cohorts with whole genome and exome sequencing, we trained and show highly accurate imputation models for MICA, MICB, HLA-E, HLA-F and HLA-G alleles base on genome SNP array data".

This limitation and the state of imputation of these gene alleles using whole genome SNP arrays like Affy PMRA or illunina infinium could have been discussed (or attempted since FinnGen has been genotyped with these). My recollection is that only Deep-HLA a deep learning based method developed by Naito and Okada in japan (Nature Communications volume 12, Article number: 1639 (2021)) exists but I am not sure how good it is for these loci. Finally, some of these genes (MICA, etc) have rare duplication and deletions, see Front. Immunol., 14 November 2023, duplication/deletions could be also mentioned as a limitation as I am not sure how these are handled.

Reviewer #2: Comments to the Authors:

In this manuscript (Tammi et al.), the authors implemented imputation models for MICA, MICB, HLA-E, HLA-F and HLA-G alleles in HIBAG framework. They used clinical-grade allelic information for Finnish dataset and also WES-based estimated alleles for 1KG dataset. They showed accuracy metrics for each of the models combining both sources. These models are publicly released through github. I think the models for imputing the non-classical genes within MHC would be valuable resource for the community. I have several major comments mostly for the metrics to assess the imputation performance, and for clarity of the description of imputation and validation methods.

1. The ‘accuracy’ metrics that the authors described are known to be upwardly biased for rare alleles by random chance. I would like to see the dosage correlation of the true alleles vs. imputed alleles as another way to assess the imputation performance, especially in the section of using 1KG as a reference model and Finnish data as the target of imputation and comparison with the clinical-grade ‘true’ alleles. It would also be useful to assess the dosage correlation as a function of minor allele frequency, which I believe is the standard of the field.

2. In the Result section (main text), I would recommend the authors to describe briefly about the Finnish dataset, including how the alleles were called, before moving onto the ‘Imputation model development’ section for readers to understand the overview of the samples used in the reference. Also, briefly describing the SNPs in the 1KG and Finnish datasets would be helpful.

3. Has the imputation performance been compared between flanking regions 50kb vs. 500kb (default of HIBAG)?

4. Considering realistic setting of the MHC imputation, it would be nice to assess the imputation performance using SNPs from commonly used genotyping array (e.g., Illumina MEGA/GSA). Using the SNPs from 1KG, the authors could create ‘mock’ genotyping data for these arrays (target) and assess the performance using Finnish-based model as a reference.

5. As another consideration for realistic setting of the MHC imputation, most of the investigators would be interested in fine-mapping the diseases alleles in the entire MHC region, not limiting to the MICA, MICB, HLA-E, HLA-F and HLA-G alleles. If the entire MHC region is imputed altogether (including classical HLA alleles), we can perform conditional analyses to identify the exhaustive combination of the MHC alleles independently associated with the disease. To this end, the reference models not only for the genes the authors focused on but also for the classical HLA alleles with the same reference samples should be very valuable. Would it be possible for the authors to release the models for the classical HLA alleles, or is it already publicly accessible?

**********

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

**********

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: Yes: Emmanuel Mignot

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

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 Ms Tammi,

We are pleased to inform you that your manuscript 'Accurate multi-population imputation of MICA, MICB, HLA-E, HLA-F and HLA-G alleles from genome SNP data' 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,

Stacey D. Finley, Ph.D.

Section Editor

PLOS Computational Biology

Stacey Finley

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 #1: I have no comments on this revision.

Reviewer #2: I appreciate the author's thorough response and additional analyses based on my previous comments. I liked the revised manuscript including additional description of the markers and the SNP-array based evaluation. It was also nice to see high dosage correlation for most common alleles. I still think that combined reference panel for the paper's genes with conventional classical HLA genes in the same set of samples is most accurate for users aiming for joint fine-mapping, rather than separate panels and post-hoc integration after imputation as the authors mentioned. But I agree that this might be beyond the scope of this manuscript.

**********

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

**********

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