蛋白质中残基远程相互作用预测算法研究综述

发布时间：2018-01-14 07:39

本文关键词：蛋白质中残基远程相互作用预测算法研究综述　出处：《计算机研究与发展》2017年01期 　论文类型：期刊论文

【摘要】：蛋白质是由多个氨基酸残基顺序连接而成的长链.在天然状态下,蛋白质并不是无规则的自由状态,而是自发形成特定的空间结构,以执行其特定的生物学功能.驱动蛋白质形成特定空间结构的主要因素是残基间的非共价相互作用,包括疏水作用、静电相互作用、范德华力等.因此,对残基之间远程相互作用的准确预测将有助于对蛋白质空间结构的预测,进而有助于对蛋白质生物学功能的了解.在蛋白质进化过程,有相互作用残基对之间存在一种"共进化"模式,即当一个残基发生变异时,与其有相互作用的残基也要发生相应的变异,以维持相互作用,进而维持整体空间结构以及生物学功能.基于上述生物学观察,研究者开发了多个统计模型和算法以预测残基对之间的相互作用:1)概述残基之间远程相互作用的两大类基本预测算法,包括无监督学习方法和监督学习方法;2)使用蛋白质结构预测CASP比赛结果来客观比较上述各类算法的性能,分析各个算法的特点和优势;3)从生物学观察和统计模型2个角度分析总结了未来的发展趋势.
[Abstract]:Protein is a long chain composed of several amino acid residues. In the natural state, protein is not an irregular free state, but spontaneous formation of a specific spatial structure. The main factors driving the formation of specific spatial structure of proteins are non-covalent interactions between residues, including hydrophobic interactions, electrostatic interactions, van der Waals forces, and so on. Accurate prediction of remote interactions between residues will help to predict the spatial structure of proteins and thus help to understand the biological functions of proteins. There is a "coevolution" pattern between the interacting residue pairs, that is, when a residue mutates, the interacting residue also has to mutate accordingly to maintain the interaction. Thus maintaining the overall spatial structure and biological functions. Based on the above biological observations. Researchers have developed a number of statistical models and algorithms to predict the interaction between residue pairs. (1) Overview of the two basic prediction algorithms for remote interactions between residues. Including unsupervised learning methods and supervised learning methods; 2) using protein structure to predict the results of CASP competition to objectively compare the performance of the above algorithms, and analyze the characteristics and advantages of each algorithm; 3) the development trend in the future is analyzed and summarized from two aspects of biological observation and statistical model.
【作者单位】：中国科学院计算技术研究所;中国科学院大学;北京大学定量生物学中心;北京大学数学科学学院;北京大学统计科学中心;中国科学院理论物理研究所;
【基金】：国家“九七三”重点基础研究发展计划基金项目(2012CB316502,2015CB910303) 国家自然科学基金项目(11175224,11121403,31270834,61272318,31171262,31428012,31471246) 中国科学院理论物理研究所理论物理国家重点实验室开放工程项目(Y4KF171CJ1)~~
【分类号】：Q811.4;TP181
【正文快照】： This work was supported by the National Basic Research Program of China(973Program)(2012CB316502,2015CB910303),theNational Natural Science Foundation of China(11175224,11121403,31270834,61272318,31171262,31428012,31471246),and the Open Project Program of

【相似文献】