[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119004-en":3,"doc-seo-119004-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},119004,549758252649,"Ivy","https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819",8,"Research & Report","Modeling Chemical Processes in Explicit Solvents with Machine Learning Potentials - Peer Review Comments","Peer review comments assess a manuscript proposing machine learning potentials for modeling chemical processes in solution. The approach uses active learning by adding new configurations selected via positions in SOAP (molecular-descriptor) space, tested on a water box and then applied to the Diels-Alder reaction between cyclopentadiene and methyl vinyl ketone in water and methanol. Reviewers request better sampling along the full reaction path, evaluation on more complex solvent-influenced systems, and longer validation MLP-MD runs with clarified NVT versus NVE ensembles. Additional concerns cover reference-accuracy and training/validation set sizes.","Peer Review File  \nModeling Chemical Processes in Explicit Solvents with Machine Learning Potentials  \nOpen Access This file is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to  \nthe original author(s) and the source, provide a link to the Creative Commons  \nlicense, and indicate if changes were made. In the cases where the authors are anonymous, such as is the case for the reports of anonymous peer reviewers, author attribution should be to 'Anonymous Referee'followed by a clear attribution to the source work. The images or other third party material in this file are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit [http://creativecommons.org/licenses/by/4.0/](http://creativecommons.org/licenses/by/4.0/) .  \nREVIEWER COMMENTS  \nReviewer \\#1 (Remarks to the Author) :  \nThis paper reports a novel approach for generating machine learning potentials (MLPs) to model chemical processes in solution. The strategy proposed relays on active learning, where new configurations are added based on their position in the molecular-descriptors (i.e. SOAP) space. This feature allows to span more efficiently the chemical and conformational space at low computational cost. Two descriptors were tested first by investigating a simple water box, to determine the quality of the potential and data efficiency. Then the strategy has been applied to study the Diels-Alder (DA) reaction between cyclopentadiene and methyl vinyl ketone in two different solvents: water and methanol. The manuscript is well written the issue well defined and both methods and results are well explained.  \nHowever, there are some concerns that need to be addressed prior to publication:  \n-The first major one regards the ability of the strategy adopted to sample the relevant chemical and conformational space. The active learning strategy adopted is based on short MD simulations (max 5 ps) using the first version of the trained MLP. These are started from configurations already present in the training set containing reactants, products, and transition state. As a consequence, I assume the MLP to be very good at modelling the three states. However, I would like to see if good sampling has been obtained also along the reaction path connecting the three.  \n-In addition, I would like to ask the authors to comment on the ability of the present strategy to model more complex system, where different paths may be activated by the presence of the solvent molecules. Would the same strategy be still effective? Or would it be necessary to include enhanced sampling techniques in the active learning strategy to explore effectively the relevant chemical space?  \n-The second concern regards the validation of the ML potential trained to model the DA reaction. This is based on a MLP-MD simulation of a box containing the substrate and 55 water molecules of  \n500 fs. The limited time makes me wonder if the MLP is actually able to reproduce therearrangement of the H-bonds in the solvent around the substrate along the reaction path and correctly account for their contribution to the reactive process. Therefore, in my opinion, longer simulations need to be performed to ascertain the ability of the potential to correctly model the dynamics of the solvent + substrate system.  \n-In addition, it is not specified whether the MLP-MD simulations used to validate the potential are performed in the NVT or NVE ensemble. Since entropy is a major player in this reaction, I think it is important to determine if the thermal fluct","cbCaignbShGfZlAS","https://ap.wps.com/l/cbCaignbShGfZlAS","pdf",3314546,1,41,"English","en",105,"# Reviewer Comments\n## Reviewer #1 (Remarks to the Author)\n## Reviewer #2 (Remarks to the Author)\n## Key Concerns Raised","[{\"question\":\"How does the proposed method generate new data for the machine learning potential?\",\"answer\":\"It uses active learning, adding configurations based on their location in molecular-descriptor space (SOAP). New structures are generated from short MD runs driven by the first trained MLP.\"},{\"question\":\"What concerns are raised about sampling and the reaction pathway?\",\"answer\":\"Reviewer #1 questions whether short MD trajectories starting from reactant/product/transition-state configurations adequately sample the connecting reaction path. The reviewer also asks whether the method can capture effects of solvent-driven activation of different paths.\"},{\"question\":\"How is the machine learning potential validated, and what validation issues are raised?\",\"answer\":\"Validation is described using MLP-MD on a box containing substrate plus explicit water for about 500 fs, but reviewers request longer simulations to ensure correct solvent/substrate dynamics. It is also unclear whether validation was done in NVT or NVE, which matters because entropy and thermal fluctuations are important.\"}]","Modeling Chemical Processes in Explicit Solvents with Machine Learning Potentials - Peer Review Comments | PDF",1785721690,103,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"modeling-chemical-processes-in-explicit-solvents-with-machine-learning-potentials-peer-review-comments","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/modeling-chemical-processes-in-explicit-solvents-with-machine-learning-potentials-peer-review-comments/119004/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"How does the proposed method generate new data for the machine learning potential?","Question",{"text":75,"@type":76},"It uses active learning, adding configurations based on their location in molecular-descriptor space (SOAP). New structures are generated from short MD runs driven by the first trained MLP.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What concerns are raised about sampling and the reaction pathway?",{"text":80,"@type":76},"Reviewer #1 questions whether short MD trajectories starting from reactant/product/transition-state configurations adequately sample the connecting reaction path. The reviewer also asks whether the method can capture effects of solvent-driven activation of different paths.",{"name":82,"@type":73,"acceptedAnswer":83},"How is the machine learning potential validated, and what validation issues are raised?",{"text":84,"@type":76},"Validation is described using MLP-MD on a box containing substrate plus explicit water for about 500 fs, but reviewers request longer simulations to ensure correct solvent/substrate dynamics. It is also unclear whether validation was done in NVT or NVE, which matters because entropy and thermal fluctuations are important.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]