Novel machine learning approaches revolutionize protein knowledge.
Novel machine learning approaches revolutionize protein knowledge.
复制标题
新颖的机器学习方法彻底改变了蛋白质知识。
DOI：
10.1016/j.tibs.2022.11.001
复制
发表时间：
2023-04
影响因子：
13.8
通讯作者：
Orengo, Christine
中科院分区：
文献类型：
--
作者：
Bordin, Nicola;Dallago, Christian;Heinzinger, Michael;Kim, Stephanie;Littmann, Maria;Rauer, Clemens;Steinegger, Martin;Rost, Burkhard;Orengo, Christine
两种基于人工智能 (AI) 的蛋白质结构预测方法 AlphaFold 2 和 RoseTTAFold 显着提高了序列结构建模的质量，接近实验精度。蛋白质语言模型对蛋白质的书面语言进行编码，与基于同源性的方法相比，可以进行更准确的注释和预测。大多数模型生物、被忽视的疾病病原体和带有精选注释的蛋白质都具有不同质量的可用模型，有助于针对单一问题问题的湿实验室实验。超快比对工具可以通过序列和结构遍历蛋白质空间，以识别以前较旧且较慢的方法无法实现的远程进化关系。对来自 21 种模型生物体的预测 AlphaFold 2 3D 模型的初步分析表明，蛋白质中的大多数 (>90%) 球状结构域可归属于当前表征的结构域进化超家族。机器学习 (ML)、蛋白质结构预测和新型超快结构对准器方面的突破性方法正在彻底改变结构生物学。获得准确的蛋白质模型并大规模注释其功能不再受到时间和资源的限制。在结构预测关键评估 (CASP) 评估中名列前茅的最新方法 AlphaFold 2 (AF2) 能够构建精度与实验结构相当的结构模型。由于蛋白质语言模型 (pLM) 和结构对准器的进步有助于验证这些转移的注释，3D 模型的注释与结构的沉积保持同步。在这篇综述中，我们描述了蛋白质科学机器学习的最新发展如何使大规模结构生物信息学可供广大科学界使用。
Two artificial intelligence (AI)-based methods for protein structure prediction, AlphaFold 2 and RoseTTAFold, increase dramatically the quality of structural modeling from sequence, nearing experimental accuracy. Protein language models encode the written language of proteins, allowing for more accurate annotations and predictions than homology-based methods. Most model organisms, neglected disease pathogens, and proteins with curated annotations have models available with varying quality, aiding wet-laboratory experiments targeting single-question issues. Ultrafast alignment tools can traverse the protein space by both sequence and structure to identify remote evolutionary relations previously precluded to older and slower methods. Preliminary analyses of predicted AlphaFold 2 3D-models from 21 model organisms suggest that the majority (>90%) of globular domains in proteins can be assigned to currently characterized domain evolutionary superfamilies. Breakthrough methods in machine learning (ML), protein structure prediction, and novel ultrafast structural aligners are revolutionizing structural biology. Obtaining accurate models of proteins and annotating their functions on a large scale is no longer limited by time and resources. The most recent method to be top ranked by the Critical Assessment of Structure Prediction (CASP) assessment, AlphaFold 2 (AF2), is capable of building structural models with an accuracy comparable to that of experimental structures. Annotations of 3D models are keeping pace with the deposition of the structures due to advancements in protein language models (pLMs) and structural aligners that help validate these transferred annotations. In this review we describe how recent developments in ML for protein science are making large-scale structural bioinformatics available to the general scientific community.
登录
查看更多内容
影响因子：
6.8
作者：
Binder, Jessica L.;Berendzen, Joel;Stevens, Amy O.;He, Yi;Wang, Jian;Dokholyan, Nikolay, V;Oprea, Tudor, I
通讯作者：
Oprea, Tudor, I
人α7烟碱乙酰胆碱受体的高度柔性细胞内结构域的结构。
影响因子：
16.6
作者：
Bondarenko V;Wells MM;Chen Q;Tillman TS;Singewald K;Lawless MJ;Caporoso J;Brandon N;Coleman JA;Saxena S;Lindahl E;Xu Y;Tang P
通讯作者：
Tang P
影响因子：
2.9
作者：
Del Alamo D;Govaerts C;Mchaourab HS
通讯作者：
Mchaourab HS
影响因子：
14.9
作者：
wwPDB consortium
通讯作者：
wwPDB consortium
影响因子：
5.8
作者：
Durairaj, Janani;Akdel, Mehmet;van Dijk, Aalt D. J.
通讯作者：
van Dijk, Aalt D. J.