MG-DIFF: A novel molecular graph diffusion model for molecular generation and optimization

Xiaochen Zhang; Shuangxi Wang; Ying Fang; Qiankun Zhang

doi:10.1371/journal.pone.0331450

Peer Review History

Original SubmissionNovember 12, 2024
4 Apr 2025 Decision Letter - Soumendranath Bhakat, Editor Dear Dr. Zhang, Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’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. Thank you for submitting your manuscript, "MG-DIFF: A Novel Molecular Graph Diffusion Model for Molecular Generation and Optimization," to PLOS ONE. The reviewers were very positive about your work but have also identified some important areas for improvement. We would be happy to consider a resubmission once you have addressed the reviewers' concerns and questions. Please ensure that your resubmission includes a link to a GitHub account hosting your code, along with a Jupyter notebook that provides clear examples and a detailed explanation of your codebase. I encourage you to carefully consider the reviewers' feedback and look forward to your revised submission. Best regards, Dr. Soumendranath Bhakat Please submit your revised manuscript by May 19 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 plosone@plos.org . When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. 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, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter. If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols . We look forward to receiving your revised manuscript. Kind regards, Soumendranath Bhakat Academic Editor PLOS ONE Journal Requirements: 1. When submitting your revision, we need you to address these additional requirements. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf 2. We suggest you thoroughly copyedit your manuscript for language usage, spelling, and grammar. If you do not know anyone who can help you do this, you may wish to consider employing a professional scientific editing service. The American Journal Experts (AJE) (https://www.aje.com/) is one such service that has extensive experience helping authors meet PLOS guidelines and can provide language editing, translation, manuscript formatting, and figure formatting to ensure your manuscript meets our submission guidelines. Please note that having the manuscript copyedited by AJE or any other editing services does not guarantee selection for peer review or acceptance for publication. Upon resubmission, please provide the following: - The name of the colleague or the details of the professional service that edited your manuscript - A copy of your manuscript showing your changes by either highlighting them or using track changes (uploaded as a supporting information file) - A clean copy of the edited manuscript (uploaded as the new manuscript file)” 3. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match. When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section. 4. Thank you for stating the following financial disclosure: “Key scientific research projects in higher education institutions of Henan Province(24A520036)” Please state what role the funders took in the study. If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript." If this statement is not correct you must amend it as needed. Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf. 5. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information . Additional Editor Comments: Dear Dr. Zhang, Thank you for submitting your manuscript, "MG-DIFF: A Novel Molecular Graph Diffusion Model for Molecular Generation and Optimization," to PLOS ONE. The reviewers were very positive about your work but have also identified some important areas for improvement. We would be happy to consider a resubmission once you have addressed the reviewers' concerns and questions. Please ensure that your resubmission includes a link to a GitHub account hosting your code, along with a Jupyter notebook that provides clear examples and a detailed explanation of your codebase. I encourage you to carefully consider the reviewers' feedback and look forward to your revised submission. Best regards, Dr. Soumendranath Bhakat [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. Is the manuscript technically sound, and do the data support the conclusions? Reviewer #1: Yes Reviewer #2: Partly ******** 2. Has the statistical analysis been performed appropriately and rigorously? -->?> Reviewer #1: Yes Reviewer #2: No ****** 3. Have the authors made all data underlying the findings in their manuscript fully available??> The PLOS Data policy Reviewer #1: Yes Reviewer #2: No ****** 4. Is the manuscript presented in an intelligible fashion and written in standard English??> Reviewer #1: Yes Reviewer #2: Yes ****** Reviewer #1: The manuscript presents MG-DIFF, a molecular graph diffusion model aimed to improve molecular generation and optimization. The authors build upon existing denoising diffusion models, incorporate a mask-and-replace discrete diffusion strategy, a graph transformer model with random node initialization, and a graph padding strategy to improve conditional generation and molecular optimization. The results show the model's performance over existing benchmarks in generating diverse molecular structures. I recommend this manuscript for publication with minor revisions. Introduction 1. The introduction provides background on existing molecular generation models and their limitations, but it would benefit from a clearer emphasis on what makes MG-DIFF distinct from prior approaches. Could the authors also highlight how their mask-and-replace strategy differs from similar techniques in previous models? A brief comparison to DiGress or other recent diffusion models in terms of efficiency, accuracy, or scalability would help contextualize MG-DIFF's unique contributions. 2. The proposed graph padding strategy for molecular optimization is intriguing, but could the authors elaborate on why padding nodes - rather than alternative graph normalization or resizing methods - are particularly effective in ensuring conditional generation? A brief mention of potential trade-offs (e.g., computational complexity and handling of small molecules) would support the rationale for this approach. Results 1. The results state that MG-Diff achieves strong performance in validity while maintaining a high internal diversity score. However, were there any observed biases or trends in the types of molecules generated? For example, does the model favor certain scaffolds or functional groups over others, and if so, could this impact the generalizability of the generated molecules? 2. While the MAD values indicate some dispersion in the conditionally generated molecular properties, could the authors provide further insight into potential causes? For example, does the noise scheduler contribute to deviations, or does the model struggle more with certain properties like TPSA than others? A brief discussion of how future improvements might refine property conditioning could be valuable. 3. The discussion on molecular optimization highlights that MG-Diff is currently limited to adding atoms but not removing them, which can constrain property refinement. Have the authors considered integrating fragment-based approaches or additional graph-editing mechanisms to allow both the addition and deletion of molecular features? Discussion and Methods 1. The discussion highlights MG-DIFF's strong performance in molecular generation and optimization. Could the authors comment on how well the model generalizes to real-world molecular datasets beyond ZINC-250K and MOSES? Would additional fine-tuning or transfer learning be required for broader applicability? 2. MG-DIFF introduces a masked and replace diffusion strategy to improve efficiency, but how does its computational cost compare to other generative models like VAEs, GANs, or traditional autoregressive approaches? A brief discussion on training time and hardware requirements would provide a better understanding of its practicality? 3. The mask-and-replace diffusion strategy is designed to deal with issues concerning uniform noise corruption, but how sensitive is the model to the choice of masking and replacement probabilities? Did the authors experiment with different noise schedules, and if so, how did they affect performance? Reviewer #2: See attached review for comments on the manuscript. While MG-DIFF achieves comparable performance on molecular generation benchmark on two datasets, further clarification is needed about the methodology and appropriate comparisons to prior work. ****** 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.] 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. Registration is free. 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 PLOS at figures@plos.org . Please note that Supporting Information files do not need this step. Attachments Attachment Submitted filename: 2025-plos-mgdiff.pdf https://doi.org/10.1371/journal.pone.0331450.r001
Revision 1
3 Jul 2025 Author Response Response to Reviewers Manuscript ID: PONE-D-24-51267 Title: MG-DIFF: A Novel Molecular Graph Diffusion Model for Molecular Generation and Optimization Dear Dr. Soumendranath Bhakat and Reviewers, We sincerely thank you for your careful review and valuable feedback on our manuscript. We appreciate the opportunity to revise and improve our work for potential publication in PLOS ONE. Below, we provide a detailed point-by-point response to all reviewer and editorial comments. All changes have been made in the revised manuscript accordingly. A marked-up version of the revised manuscript is also provided for your convenience. We have carefully revised our paper to meet the requirements. If there are other quetions, please let ours know. The manuscript has been professionally copyedited by our college Yifan Dong. We have improved grammar, clarity, and consistency throughout. We have corrected the grant information. The following statement has also been added as requested:“The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.”We also added supporting information captians as needed. We sincerely thank you for the valuable time and effort all editor and revievers have dedicated to reviewing our manuscript. We are especially grateful for the constructive and insightful comments provided, which have significantly helped us improve the quality and clarity of our manuscript. Below we provide a point-by-point response to each comment. All changes have been incorporated in the revised manuscript, and we have highlighted the modifications where appropriate. Reviewer #1: 1. The introduction provides background on existing molecular generation models and their limitations, but it would benefit from a clearer emphasis on what makes MG-DIFF distinct from prior approaches. Could the authors also highlight how their mask-and-replace strategy differs from similar techniques in previous models? A brief comparison to DiGress or other recent diffusion models in terms of efficiency, accuracy, or scalability would help contextualize MG-DIFF's unique contributions. Reply: Thank you for this valuable suggestion. In the revised Introduction (Paragraphs 7–8), we have explicitly highlighted the differences between MG-DIFF and prior diffusion models, especially DiGress. We clarify that MG-DIFF introduces (1) a novel mask-and-replace discrete diffusion strategy, which improves convergence and denoising stability compared to the aggressive uniform noise used in DiGress, (2) a graph transformer with random node initialization to improve expressiveness beyond the first-order Weisfeiler-Lehman test, and (3) a graph padding strategy that enables flexible conditional generation and optimization. 2. The proposed graph padding strategy for molecular optimization is intriguing, but could the authors elaborate on why padding nodes - rather than alternative graph normalization or resizing methods - are particularly effective in ensuring conditional generation? A brief mention of potential trade-offs (e.g., computational complexity and handling of small molecules) would support the rationale for this approach. Reply: Unlike resizing or normalization, padding ensures consistent input dimensions while maintaining molecular semantics. The padding nodes are not bonded and can be cleanly removed post-generation. This strategy allows the model to learn the distribution of node counts and flexibly add atoms during optimization. 3. The results state that MG-Diff achieves strong performance in validity while maintaining a high internal diversity score. However, were there any observed biases or trends in the types of molecules generated? For example, does the model favor certain scaffolds or functional groups over others, and if so, could this impact the generalizability of the generated molecules? Reply: We appreciate this insightful point. Although our quantitative metrics suggest good diversity and novelty, we conducted a qualitative inspection of generated molecules and found no strong bias toward specific scaffolds. 4. While the MAD values indicate some dispersion in the conditionally generated molecular properties, could the authors provide further insight into potential causes? For example, does the noise scheduler contribute to deviations, or does the model struggle more with certain properties like TPSA than others? A brief discussion of how future improvements might refine property conditioning could be valuable. Reply:Thank you for the suggestion. In the revised Conditional Generation section, we explain that dispersion likely arises from two sources: the stochastic nature of the mask-and-replace scheduler and the inherent difficulty of tightly controlling complex properties (e.g., TPSA). We also note that while MAD indicates dispersion, the property distributions still cluster near the target values, which supports the effectiveness of our method. Future improvements could include adaptive noise scheduling or property-aware loss functions. 5. The discussion on molecular optimization highlights that MG-Diff is currently limited to adding atoms but not removing them, which can constrain property refinement. Have the authors considered integrating fragment-based approaches or additional graph-editing mechanisms to allow both the addition and deletion of molecular features? Reply: Thank you for the suggestion. In the revised Conditional Generation section, we explain that dispersion likely arises from two sources: the stochastic nature of the mask-and-replace scheduler and the inherent difficulty of tightly controlling complex properties (e.g., TPSA). We also note that while MAD indicates dispersion, the property distributions still cluster near the target values, which supports the effectiveness of our method. Future improvements could include adaptive noise scheduling or property-aware loss functions. 6. The discussion highlights MG-DIFF's strong performance in molecular generation and optimization. Could the authors comment on how well the model generalizes to real-world molecular datasets beyond ZINC-250K and MOSES? Would additional fine-tuning or transfer learning be required for broader applicability? Reply: We have added a statement in the Discussion emphasizing that while ZINC-250K and MOSES are standard benchmarks, MG-DIFF is designed to be general and modular. For domain-specific applications (e.g., ChEMBL or PubChem), fine-tuning or transfer learning would be necessary to adapt to new property distributions or chemical spaces. 7. MG-DIFF introduces a masked and replace diffusion strategy to improve efficiency, but how does its computational cost compare to other generative models like VAEs, GANs, or traditional autoregressive approaches? A brief discussion on training time and hardware requirements would provide a better understanding of its practicality? Reply: In the revised Supplementary Information, we now include a note that MG-DIFF requires moderate training time (e.g., ~20 hours on a single RTX 4090 for ZINC-250K), which is slightly more than VAE-based models but comparable to DiGress. The training benefits from efficient parallelization and stable convergence thanks to our mask-and-replace strategy. 8. The mask-and-replace diffusion strategy is designed to deal with issues concerning uniform noise corruption, but how sensitive is the model to the choice of masking and replacement probabilities? Did the authors experiment with different noise schedules, and if so, how did they affect performance? Reply: Yes, we have conducted ablation experiments with various replacement rates (0%–5%). We found that a replacement rate of 1–2% yielded the best trade-off between stability and flexibility. Further details can be find in our open-source codes. Reviewer #2: See attached review for comments on the manuscript. While MG-DIFF achieves comparable performance on molecular generation benchmark on two datasets, further clarification is needed about the methodology and appropriate comparisons to prior work. Reply: We thank the reviewer for this comment. We have revised the Introduction and Discussion to better contextualize MG-DIFF relative to DiGress, VAE-based, and autoregressive models (e.g., MolGPT). We emphasize the novelty of our mask-and-replace scheduler, transformer-based denoising network, and padding strategy, which together enable state-of-the-art performance on two benchmarks while also supporting conditional generation and optimization tasks. Once again, we sincerely thank the reviewers for their detailed feedback. We believe that the revised manuscript has significantly improved in clarity, completeness, and rigor based on your comments. Sincerely, Dr. Zhang Xiaocen, on behalf of all co-authors Shangqiu Normal University zhangxiaochen@nudt.edu.cn Attachments Attachment Submitted filename: responds.docx https://doi.org/10.1371/journal.pone.0331450.r002
17 Aug 2025 Decision Letter - Soumendranath Bhakat, Editor MG-DIFF: A Novel Molecular Graph Diffusion Model for Molecular Generation and Optimization PONE-D-24-51267R1 Dear Dr. Zhang, We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements. Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication. An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact billing support . If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. Kind regards, Soumendranath Bhakat Academic Editor PLOS ONE Additional Editor Comments (optional): Dear Dr. Zhang, Thanks for a great work on the revising the manuscript. I am happy to accept the manuscript in its updated version. Best regards, Dr. Soumendranath Bhakat Reviewers' comments: Reviewer's Responses to Questions Comments to the Author Reviewer #1: All comments have been addressed Reviewer #3: All comments have been addressed ******** 2. Is the manuscript technically sound, and do the data support the conclusions??> Reviewer #1: Partly Reviewer #3: Yes ****** 3. Has the statistical analysis been performed appropriately and rigorously? -->?> Reviewer #1: Yes Reviewer #3: Yes ****** 4. Have the authors made all data underlying the findings in their manuscript fully available??> The PLOS Data policy Reviewer #1: Yes Reviewer #3: Yes ****** 5. Is the manuscript presented in an intelligible fashion and written in standard English??> Reviewer #1: Yes Reviewer #3: Yes ****** Reviewer #1: I am satisfied with the responses along with the revision in the manuscript. I recommend publication of the manuscript. Reviewer #3: (No Response) ****** 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 #3: No ******** https://doi.org/10.1371/journal.pone.0331450.r003
Formally Accepted
Acceptance Letter - Soumendranath Bhakat, Editor PONE-D-24-51267R1 PLOS ONE Dear Dr. Zhang, I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now being handed over to our production team. At this stage, our production department will prepare your paper for publication. This includes ensuring the following: * All references, tables, and figures are properly cited * All relevant supporting information is included in the manuscript submission, * There are no issues that prevent the paper from being properly typeset You will receive further instructions from the production team, including instructions on how to review your proof when it is ready. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few days to review your paper and let you know the next and final steps. Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. You will receive an invoice from PLOS for your publication fee after your manuscript has reached the completed accept phase. If you receive an email requesting payment before acceptance or for any other service, this may be a phishing scheme. Learn how to identify phishing emails and protect your accounts at https://explore.plos.org/phishing. If we can help with anything else, please email us at customercare@plos.org. Thank you for submitting your work to PLOS ONE and supporting open access. Kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Soumendranath Bhakat Academic Editor PLOS ONE https://doi.org/10.1371/journal.pone.0331450.r004

Open letter on the publication of peer review reports

PLOS recognizes the benefits of transparency in the peer review process. Therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. Reviewers remain anonymous, unless they choose to reveal their names.

We encourage other journals to join us in this initiative. We hope that our action inspires the community, including researchers, research funders, and research institutions, to recognize the benefits of published peer review reports for all parts of the research system.

Learn more at ASAPbio .