Nataša Tagasovska, Ji Won Park, Matthieu Kirchmeyer, Nathan C. Frey, Andrew Martin Watkins, Aya Abdelsalam Ismail, Arian Rokkum Jamasb, Edith Lee, Tyler Bryson, Stephen Ra, Kyunghyun Cho
{"title":"抗体域床:治疗蛋白质设计中的分布外泛化","authors":"Nataša Tagasovska, Ji Won Park, Matthieu Kirchmeyer, Nathan C. Frey, Andrew Martin Watkins, Aya Abdelsalam Ismail, Arian Rokkum Jamasb, Edith Lee, Tyler Bryson, Stephen Ra, Kyunghyun Cho","doi":"arxiv-2407.21028","DOIUrl":null,"url":null,"abstract":"Machine learning (ML) has demonstrated significant promise in accelerating\ndrug design. Active ML-guided optimization of therapeutic molecules typically\nrelies on a surrogate model predicting the target property of interest. The\nmodel predictions are used to determine which designs to evaluate in the lab,\nand the model is updated on the new measurements to inform the next cycle of\ndecisions. A key challenge is that the experimental feedback from each cycle\ninspires changes in the candidate proposal or experimental protocol for the\nnext cycle, which lead to distribution shifts. To promote robustness to these\nshifts, we must account for them explicitly in the model training. We apply\ndomain generalization (DG) methods to classify the stability of interactions\nbetween an antibody and antigen across five domains defined by design cycles.\nOur results suggest that foundational models and ensembling improve predictive\nperformance on out-of-distribution domains. We publicly release our codebase\nextending the DG benchmark ``DomainBed,'' and the associated dataset of\nantibody sequences and structures emulating distribution shifts across design\ncycles.","PeriodicalId":501022,"journal":{"name":"arXiv - QuanBio - Biomolecules","volume":"16 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Antibody DomainBed: Out-of-Distribution Generalization in Therapeutic Protein Design\",\"authors\":\"Nataša Tagasovska, Ji Won Park, Matthieu Kirchmeyer, Nathan C. Frey, Andrew Martin Watkins, Aya Abdelsalam Ismail, Arian Rokkum Jamasb, Edith Lee, Tyler Bryson, Stephen Ra, Kyunghyun Cho\",\"doi\":\"arxiv-2407.21028\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Machine learning (ML) has demonstrated significant promise in accelerating\\ndrug design. Active ML-guided optimization of therapeutic molecules typically\\nrelies on a surrogate model predicting the target property of interest. The\\nmodel predictions are used to determine which designs to evaluate in the lab,\\nand the model is updated on the new measurements to inform the next cycle of\\ndecisions. A key challenge is that the experimental feedback from each cycle\\ninspires changes in the candidate proposal or experimental protocol for the\\nnext cycle, which lead to distribution shifts. To promote robustness to these\\nshifts, we must account for them explicitly in the model training. We apply\\ndomain generalization (DG) methods to classify the stability of interactions\\nbetween an antibody and antigen across five domains defined by design cycles.\\nOur results suggest that foundational models and ensembling improve predictive\\nperformance on out-of-distribution domains. We publicly release our codebase\\nextending the DG benchmark ``DomainBed,'' and the associated dataset of\\nantibody sequences and structures emulating distribution shifts across design\\ncycles.\",\"PeriodicalId\":501022,\"journal\":{\"name\":\"arXiv - QuanBio - Biomolecules\",\"volume\":\"16 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-07-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - QuanBio - Biomolecules\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2407.21028\",\"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 - QuanBio - Biomolecules","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2407.21028","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
机器学习(ML)已在加速药物设计方面展现出巨大前景。主动式 ML 引导的治疗分子优化通常依赖于预测相关目标特性的代理模型。模型预测用于确定在实验室中评估哪些设计,并根据新的测量结果更新模型,为下一轮决策提供信息。一个关键的挑战是,每个周期的实验反馈都会促使下一个周期的候选方案或实验方案发生变化,从而导致分布转移。为了提高对这些变化的稳健性,我们必须在模型训练中明确考虑这些变化。我们应用领域泛化(DG)方法对抗体和抗原之间的相互作用在设计周期所定义的五个领域中的稳定性进行了分类。我们的结果表明,基础模型和集合提高了对分布外领域的预测性能。我们公开发布了扩展 DG 基准 "DomainBed "的代码库,以及相关的抗体序列和结构数据集,以模拟跨设计周期的分布转移。
Antibody DomainBed: Out-of-Distribution Generalization in Therapeutic Protein Design
Machine learning (ML) has demonstrated significant promise in accelerating
drug design. Active ML-guided optimization of therapeutic molecules typically
relies on a surrogate model predicting the target property of interest. The
model predictions are used to determine which designs to evaluate in the lab,
and the model is updated on the new measurements to inform the next cycle of
decisions. A key challenge is that the experimental feedback from each cycle
inspires changes in the candidate proposal or experimental protocol for the
next cycle, which lead to distribution shifts. To promote robustness to these
shifts, we must account for them explicitly in the model training. We apply
domain generalization (DG) methods to classify the stability of interactions
between an antibody and antigen across five domains defined by design cycles.
Our results suggest that foundational models and ensembling improve predictive
performance on out-of-distribution domains. We publicly release our codebase
extending the DG benchmark ``DomainBed,'' and the associated dataset of
antibody sequences and structures emulating distribution shifts across design
cycles.