{"title":"从多个来源的深度突变扫描数据学习蛋白质适应度景观。","authors":"Lin Chen, Zehong Zhang, Zhenghao Li, Rui Li, Ruifeng Huo, Lifan Chen, Dingyan Wang, Xiaomin Luo, Kaixian Chen, Cangsong Liao, Mingyue Zheng","doi":"10.1016/j.cels.2023.07.003","DOIUrl":null,"url":null,"abstract":"<p><p>One of the key points of machine learning-assisted directed evolution (MLDE) is the accurate learning of the fitness landscape, a conceptual mapping from sequence variants to the desired function. Here, we describe a multi-protein training scheme that leverages the existing deep mutational scanning data from diverse proteins to aid in understanding the fitness landscape of a new protein. Proof-of-concept trials are designed to validate this training scheme in three aspects: random and positional extrapolation for single-variant effects, zero-shot fitness predictions for new proteins, and extrapolation for higher-order variant effects from single-variant effects. Moreover, our study identified previously overlooked strong baselines, and their unexpectedly good performance brings our attention to the pitfalls of MLDE. Overall, these results may improve our understanding of the association between different protein fitness profiles and shed light on developing better machine learning-assisted approaches to the directed evolution of proteins. A record of this paper's transparent peer review process is included in the supplemental information.</p>","PeriodicalId":54348,"journal":{"name":"Cell Systems","volume":null,"pages":null},"PeriodicalIF":9.0000,"publicationDate":"2023-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Learning protein fitness landscapes with deep mutational scanning data from multiple sources.\",\"authors\":\"Lin Chen, Zehong Zhang, Zhenghao Li, Rui Li, Ruifeng Huo, Lifan Chen, Dingyan Wang, Xiaomin Luo, Kaixian Chen, Cangsong Liao, Mingyue Zheng\",\"doi\":\"10.1016/j.cels.2023.07.003\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>One of the key points of machine learning-assisted directed evolution (MLDE) is the accurate learning of the fitness landscape, a conceptual mapping from sequence variants to the desired function. Here, we describe a multi-protein training scheme that leverages the existing deep mutational scanning data from diverse proteins to aid in understanding the fitness landscape of a new protein. Proof-of-concept trials are designed to validate this training scheme in three aspects: random and positional extrapolation for single-variant effects, zero-shot fitness predictions for new proteins, and extrapolation for higher-order variant effects from single-variant effects. Moreover, our study identified previously overlooked strong baselines, and their unexpectedly good performance brings our attention to the pitfalls of MLDE. Overall, these results may improve our understanding of the association between different protein fitness profiles and shed light on developing better machine learning-assisted approaches to the directed evolution of proteins. A record of this paper's transparent peer review process is included in the supplemental information.</p>\",\"PeriodicalId\":54348,\"journal\":{\"name\":\"Cell Systems\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":9.0000,\"publicationDate\":\"2023-08-16\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Cell Systems\",\"FirstCategoryId\":\"99\",\"ListUrlMain\":\"https://doi.org/10.1016/j.cels.2023.07.003\",\"RegionNum\":1,\"RegionCategory\":\"生物学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"BIOCHEMISTRY & MOLECULAR BIOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Cell Systems","FirstCategoryId":"99","ListUrlMain":"https://doi.org/10.1016/j.cels.2023.07.003","RegionNum":1,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"BIOCHEMISTRY & MOLECULAR BIOLOGY","Score":null,"Total":0}
Learning protein fitness landscapes with deep mutational scanning data from multiple sources.
One of the key points of machine learning-assisted directed evolution (MLDE) is the accurate learning of the fitness landscape, a conceptual mapping from sequence variants to the desired function. Here, we describe a multi-protein training scheme that leverages the existing deep mutational scanning data from diverse proteins to aid in understanding the fitness landscape of a new protein. Proof-of-concept trials are designed to validate this training scheme in three aspects: random and positional extrapolation for single-variant effects, zero-shot fitness predictions for new proteins, and extrapolation for higher-order variant effects from single-variant effects. Moreover, our study identified previously overlooked strong baselines, and their unexpectedly good performance brings our attention to the pitfalls of MLDE. Overall, these results may improve our understanding of the association between different protein fitness profiles and shed light on developing better machine learning-assisted approaches to the directed evolution of proteins. A record of this paper's transparent peer review process is included in the supplemental information.
Cell SystemsMedicine-Pathology and Forensic Medicine
CiteScore
16.50
自引率
1.10%
发文量
84
审稿时长
42 days
期刊介绍:
In 2015, Cell Systems was founded as a platform within Cell Press to showcase innovative research in systems biology. Our primary goal is to investigate complex biological phenomena that cannot be simply explained by basic mathematical principles. While the physical sciences have long successfully tackled such challenges, we have discovered that our most impactful publications often employ quantitative, inference-based methodologies borrowed from the fields of physics, engineering, mathematics, and computer science. We are committed to providing a home for elegant research that addresses fundamental questions in systems biology.