Interpretable Integration of Unpaired Multi-Omics for Alzheimer’s Diagnosis via Cross-Modal Transformer Reconstruction

PLOS Computational Biology

Dear Dr. Li,

Thank you for submitting your manuscript to PLOS Computational Biology. After careful consideration, we feel that it has merit but does not fully meet PLOS Computational Biology's publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript within 60 days Dec 21 2025 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at ploscompbiol@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pcompbiol/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

* A rebuttal letter that responds to each point raised by the editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. This file does not need to include responses to formatting updates and technical items listed in the 'Journal Requirements' section below.

* A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

* An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, competing interests statement, or data availability statement, please make these updates within the submission form at the time of resubmission. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter

We look forward to receiving your revised manuscript.

Kind regards,

Samuel V. Scarpino

Academic Editor

PLOS Computational Biology

Shihua Zhang

Section Editor

PLOS Computational Biology

Additional Editor Comment:

I agree with the reviewers that the manuscript presents an interesting approach to explainability that could potentially be of widespread appeal in the community. However, I also agree with the reviewers that strong evidence exists that the transformer-based method presented as the backbone of the paper is overfitting. The reviewers cite two pieces of evidence in their assessment. First, we have an a priori expectation that a transformer model might overfit to the data given the vastly larger number of parameters it likely contains (as compared to other comparator methods used in the paper). Second, we can see in Table 1 that performance of the transformer-based model falls off more on the external test set than we see in the other models. That being said, it is quite interesting that the transformer-based model still seems to do the best on--as least based on accuracy--the external test set (all models struggled). I also agree with the reviewer comments that more works is needed to justify the biological relevance of the results given that the input data was not paired at the individual-level (I don't believe this is a fatal flaw, but one that should be discussed in more detail and possibly explored with sensitivity analyses using a simulation model). In addition to the reviewer comments, which I encourage the authors to pay careful attention to in their revision, I think the following additional items should be considered:

1. In Table 1, please report precision and recall in addition to accuracy (or a similar set of measures). I would also expect to see the baseline expectation reported for random guessing. It appears as though the sample sizes are the same across all models, so that column could be replaced and the sample sizes provided in a table caption. It would also be quite helpful to include an additional column listing the number of parameters in each model.

1b. I also think that Table 1 is quite hard to parse. Please consider adding shading or lines so that the reader can more quickly assess differences. You might also consider re-ordering the models such that they are presented in descending order based on external test-set performance. Be sure in the legend to explain the table fully, including how ordering was determined

2. The figure and table captions should provide more information on how to interpret the figure without the need to refer back to the text. In some cases, the titles are also too colloquial, e.g., F4 says "Accuracy..." but then many other measures of performance are provided. Similar issues appear in other figure titles. This and the above point may seem like nitpicking, but I believe it's easy to misinterpret the findings (especially when readers are skimming).

3. For Figures 4 and other generalizability results (e.g., 5 and 6), I would expect to see features included from one of the other top performing models. Presumably you could use the same feature selection procedure (showing how your approach to explainability generalizes to other complicated model structures would strengthen the paper significantly). The best solution here would be to add an additional panel with all the models in Table 1. If the authors believe that is too busy, then you could add just a single model (maybe the deep belief or LR) and then add a supplemental figure showing all the others.

I thank the authors for sending us their manuscript and will look forward to receiving a revision.

Journal Requirements:

1) We ask that a manuscript source file is provided at Revision. Please upload your manuscript file as a .doc, .docx, .rtf or .tex. If you are providing a .tex file, please upload it under the item type u2018LaTeX Source Fileu2019 and leave your .pdf version as the item type u2018Manuscriptu2019.

2) Please provide an Author Summary. This should appear in your manuscript between the Abstract (if applicable) and the Introduction, and should be 150-200 words long. The aim should be to make your findings accessible to a wide audience that includes both scientists and non-scientists. Sample summaries can be found on our website under Submission Guidelines:

https://journals.plos.org/ploscompbiol/s/submission-guidelines#loc-parts-of-a-submission

3) Please upload all main figures as separate Figure files in .tif or .eps format. For more information about how to convert and format your figure files please see our guidelines:

https://journals.plos.org/ploscompbiol/s/figures

4) Please ensure that all Figure files have corresponding citations and legends within the manuscript. Currently, Figures 2, and 7 in your submission file inventory do not have in-text citations. Please include the in-text citations of the figures.

5) We have noticed that you have uploaded Supporting Information files, but you have not included a list of legends. Please add a full list of legends for your Supporting Information files after the references list.

6) Some material included in your submission may be copyrighted. According to PLOSu2019s copyright policy, authors who use figures or other material (e.g., graphics, clipart, maps) from another author or copyright holder must demonstrate or obtain permission to publish this material under the Creative Commons Attribution 4.0 International (CC BY 4.0) License used by PLOS journals. Please closely review the details of PLOSu2019s copyright requirements here: PLOS Licenses and Copyright. If you need to request permissions from a copyright holder, you may use PLOS's Copyright Content Permission form.

Please respond directly to this email and provide any known details concerning your material's license terms and permissions required for reuse, even if you have not yet obtained copyright permissions or are unsure of your material's copyright compatibility. Once you have responded and addressed all other outstanding technical requirements, you may resubmit your manuscript within Editorial Manager.

Potential Copyright Issues:

i) Figure 1A. Please confirm whether you drew the images / clip-art within the figure panels by hand. If you did not draw the images, please provide (a) a link to the source of the images or icons and their license / terms of use; or (b) written permission from the copyright holder to publish the images or icons under our CC BY 4.0 license. Alternatively, you may replace the images with open source alternatives. See these open source resources you may use to replace images / clip-art:

- https://commons.wikimedia.org

- https://openclipart.org/.

7) Please amend your detailed Financial Disclosure statement. This is published with the article. It must therefore be completed in full sentences and contain the exact wording you wish to be published.

1) State the initials, alongside each funding source, of each author to receive each grant. For example: "This work was supported by the National Institutes of Health (####### to AM; ###### to CJ) and the National Science Foundation (###### to AM)."

2) State what role the funders took in the study. If the funders had no role in your study, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

8) Your current Financial Disclosure states receiving several funds. However, your funding information on the submission form doesn't indicate any funds. Please ensure that the funders and grant numbers match between the Financial Disclosure field and the Funding Information tab in your submission form. Note that the funders must be provided in the same order in both places as well.

9) Please revise your current Competing Interest statement to the standard "The authors have declared that no competing interests exist."

Note: If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise.

Reviewers' comments:

Reviewer's Responses to Questions

Reviewer #1: This manuscript presents AE-Trans, a dual-channel Transformer framework for integrating unpaired RNA-seq and DNA methylation data for Alzheimer's disease diagnosis. While the work addresses an important clinical need and demonstrates promising results, several methodological concerns and missing technical details limit the robustness of the findings:

- The motivation of using transformer architecture is not clear. The datasets used for the training are small for transformer model and the chance of overfitting is high.

- Does Cartesian intra-label approach truly capture inter-omics relationships as they exist in individual patients, or does it primarily learn label-specific patterns that may not generalize to real biological systems?

- It is not clear whether all baseline models (RF, NB, LR, DEG-DMP-DNN, AE-XGBoost, DBN) were trained on the same harmonized 14,926-gene feature space as AE-Trans.

- The gap in accuracy between Test (CV, Test) and external set support the idea of overfitting. This should be explained by the authors.

Reviewer #2: Your work addresses an important clinical problem and proposes an innovative technical framework that combines several promising methodologies. The work has many strengths: the interpretability framework using counterfactual integrated gradients is valuable, the identified biological pathways show reasonable concordance with AD literature, and the technical innovation of bidirectional reconstruction in Transformers has potential merit. However, there are some concerns about the experimental design and data integration strategy that must be addressed before publication:

- Creating all possible combinations between RNA-AD samples and Methylation-AD samples fundamentally violates the biological reality that each patient has unique cross-modal molecular relationships. This synthetic pairing strategy assumes all AD patients share identical RNA-methylation associations, which is biologically implausible given the known heterogeneity of Alzheimer's disease.

This approach creates several problems: 1. it generates artificial relationships that don't exist in real patients, 2. it dramatically inflates your dataset size with redundant synthetic combinations, and 3. it may explain the unusually high performance metrics that seem inconsistent with the complexity of AD diagnosis.

- Different brain regions have distinct molecular profiles and assuming AD signatures are identical across regions lacks support. This may explain the substantial performance drop in external validation.

- The results require validation in larger, independent studies before drawing clinical conclusions. The feature selection validation using the same data for both identification and testing risks overfitting.

- Several technical aspects need clarification: the "masking mechanism" mentioned in ablation studies is never properly explained, the exact implementation of your synthetic pairing during cross-validation is unclear, and batch harmonization details are insufficient for reproducibility. The sample size discrepancies between your described 80/20 split and the reported table values need resolution.

**********

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: None

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

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

Figure resubmission:

After uploading your figures to PLOS’s NAAS tool - https://ngplosjournals.pagemajik.ai/artanalysis, NAAS will process the files provided and display the results in the "Uploaded Files" section of the page as the processing is complete. If the uploaded figures meet our requirements (or NAAS is able to fix the files to meet our requirements), the figure will be marked as "fixed" above. If NAAS is unable to fix the files, a red "failed" label will appear above. When NAAS has confirmed that the figure files meet our requirements, please download the file via the download option, and include these NAAS processed figure files when submitting your revised manuscript.

Reproducibility:

To enhance the reproducibility of your results, we recommend that authors of applicable studies 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. 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

PCOMPBIOL-D-25-01596R1

Interpretable Integration of Unpaired Multi-Omics for Alzheimer’s Diagnosis via Cross-Modal Transformer Reconstruction

PLOS Computational Biology

Dear Dr. Li,

Thank you for submitting your manuscript to PLOS Computational Biology. After careful consideration, we feel that it has merit but does not fully meet PLOS Computational Biology's publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Apr 21 2026 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at ploscompbiol@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pcompbiol/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

* A letter that responds to each point raised by the editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. This file does not need to include responses to formatting updates and technical items listed in the 'Journal Requirements' section below.

* A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

* An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, competing interests statement, or data availability statement, please make these updates within the submission form at the time of resubmission. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

We look forward to receiving your revised manuscript.

Kind regards,

Samuel V. Scarpino

Academic Editor

PLOS Computational Biology

Shihua Zhang

Section Editor

PLOS Computational Biology

Additional Editor Comments:

I agree with R2 that the authors should add some additional caveats regarding the risk of overfitting.

Journal Requirements:

If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise.

1) Your manuscript is missing the following sections: Methods.  Please ensure all required sections are present and in the correct order. Make sure section heading levels are clearly indicated in the manuscript text, and limit sub-sections to 3 heading levels. An outline of the required sections can be consulted in our submission guidelines here:

https://journals.plos.org/ploscompbiol/s/submission-guidelines#loc-parts-of-a-submission

2) Please amend your detailed Financial Disclosure statement. This is published with the article. It must therefore be completed in full sentences and contain the exact wording you wish to be published.

1) State the initials, alongside each funding source, of each author to receive each grant. For example: "This work was supported by the National Institutes of Health (####### to AM; ###### to CJ) and the National Science Foundation (###### to AM)."

2) State what role the funders took in the study. If the funders had no role in your study, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

3) If any authors received a salary from any of your funders, please state which authors and which funders..

If you did not receive any funding for this study, please simply state: u201cThe authors received no specific funding for this work.u201d

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Authors:

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

Reviewer #1: The authors answered my concerns carefully.

Reviewer #2: Thank you for addressing the reviewers' comments. The risk of overfitting remains a concern given the very large model size relative to the available data. Although the added external datasets and experiments are helpful, a more cautious interpretation of the results and claims would be advised.

**********

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: None

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

Figure resubmission:

Reproducibility:

To enhance the reproducibility of your results, we recommend that authors of applicable studies 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. 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 Mr Li,

We are pleased to inform you that your manuscript 'Interpretable Integration of Unpaired Multi-Omics for Alzheimer’s Diagnosis via Cross-Modal Transformer Reconstruction' 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,

Samuel V. Scarpino

Academic Editor

PLOS Computational Biology

Shihua Zhang

Section Editor

PLOS Computational Biology

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