{"title":"几何图神经网络中的无描述符集合变量","authors":"Jintu Zhang, Luigi Bonati, Enrico Trizio, Odin Zhang, Yu Kang, TingJun Hou, Michele Parrinello","doi":"arxiv-2409.07339","DOIUrl":null,"url":null,"abstract":"Enhanced sampling simulations make the computational study of rare events\nfeasible. A large family of such methods crucially depends on the definition of\nsome collective variables (CVs) that could provide a low-dimensional\nrepresentation of the relevant physics of the process. Recently, many methods\nhave been proposed to semi-automatize the CV design by using machine learning\ntools to learn the variables directly from the simulation data. However, most\nmethods are based on feed-forward neural networks and require as input some\nuser-defined physical descriptors. Here, we propose to bypass this step using a\ngraph neural network to directly use the atomic coordinates as input for the CV\nmodel. This way, we achieve a fully automatic approach to CV determination that\nprovides variables invariant under the relevant symmetries, especially the\npermutational one. Furthermore, we provide different analysis tools to favor\nthe physical interpretation of the final CV. We prove the robustness of our\napproach using different methods from the literature for the optimization of\nthe CV, and we prove its efficacy on several systems, including a small\npeptide, an ion dissociation in explicit solvent, and a simple chemical\nreaction.","PeriodicalId":501369,"journal":{"name":"arXiv - PHYS - Computational Physics","volume":"47 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Descriptors-free Collective Variables From Geometric Graph Neural Networks\",\"authors\":\"Jintu Zhang, Luigi Bonati, Enrico Trizio, Odin Zhang, Yu Kang, TingJun Hou, Michele Parrinello\",\"doi\":\"arxiv-2409.07339\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Enhanced sampling simulations make the computational study of rare events\\nfeasible. A large family of such methods crucially depends on the definition of\\nsome collective variables (CVs) that could provide a low-dimensional\\nrepresentation of the relevant physics of the process. Recently, many methods\\nhave been proposed to semi-automatize the CV design by using machine learning\\ntools to learn the variables directly from the simulation data. However, most\\nmethods are based on feed-forward neural networks and require as input some\\nuser-defined physical descriptors. Here, we propose to bypass this step using a\\ngraph neural network to directly use the atomic coordinates as input for the CV\\nmodel. This way, we achieve a fully automatic approach to CV determination that\\nprovides variables invariant under the relevant symmetries, especially the\\npermutational one. Furthermore, we provide different analysis tools to favor\\nthe physical interpretation of the final CV. We prove the robustness of our\\napproach using different methods from the literature for the optimization of\\nthe CV, and we prove its efficacy on several systems, including a small\\npeptide, an ion dissociation in explicit solvent, and a simple chemical\\nreaction.\",\"PeriodicalId\":501369,\"journal\":{\"name\":\"arXiv - PHYS - Computational Physics\",\"volume\":\"47 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - PHYS - Computational Physics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.07339\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - PHYS - Computational Physics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.07339","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Descriptors-free Collective Variables From Geometric Graph Neural Networks
Enhanced sampling simulations make the computational study of rare events
feasible. A large family of such methods crucially depends on the definition of
some collective variables (CVs) that could provide a low-dimensional
representation of the relevant physics of the process. Recently, many methods
have been proposed to semi-automatize the CV design by using machine learning
tools to learn the variables directly from the simulation data. However, most
methods are based on feed-forward neural networks and require as input some
user-defined physical descriptors. Here, we propose to bypass this step using a
graph neural network to directly use the atomic coordinates as input for the CV
model. This way, we achieve a fully automatic approach to CV determination that
provides variables invariant under the relevant symmetries, especially the
permutational one. Furthermore, we provide different analysis tools to favor
the physical interpretation of the final CV. We prove the robustness of our
approach using different methods from the literature for the optimization of
the CV, and we prove its efficacy on several systems, including a small
peptide, an ion dissociation in explicit solvent, and a simple chemical
reaction.