计算机工程与应用 ›› 2017, Vol. 53 ›› Issue (14): 111-116.DOI: 10.3778/j.issn.1002-8331.1602-0043

• 模式识别与人工智能 • 上一篇    下一篇

面向多肽的分子对接算法性能对比研究

罗  浩1,2,刘  宇1,2,吴思晋3   

  1. 1.大连理工大学 软件学院,辽宁 大连 116024
    2.大连理工大学 IT服务工程与管理研究所,辽宁 大连 116024
    3.大连理工大学 生命科学与技术学院,辽宁 大连 116024
  • 出版日期:2017-07-15 发布日期:2017-08-01

Comprehensive comparative study of protein-peptide docking algorithms performance

LUO Hao1,2, LIU Yu1,2, WU Sijin3   

  1. 1.School of Software, Dalian University of Technology, Dalian, Liaoning 116024, China
    2.Institute of IT Service Engineering and Management, Dalian University of Technology, Dalian, Liaoning 116024, China
    3.School of Life Science and Biotechnology, Dalian University of Technology, Dalian, Liaoning 116024, China
  • Online:2017-07-15 Published:2017-08-01

摘要: 蛋白质-多肽间相互作用是计算生物学中的热门方向,对其结合的精确预测对于了解体内复杂的信号通路以及蛋白质功能,及进一步进行药物设计有着至关重要的作用。蛋白质-多肽对接方法能够有效地模拟蛋白质-多肽间相互作用,其对接性能直接决定模拟实验的准确性和成功与否。针对当前主要的蛋白质-多肽对接方法,综合测试了其算法性能;同时将本实验室开发的基于全信息粒子群算法的分子对接工具FIPSDock(Fully-Informed-Particle-Swarm-Dock)延伸到蛋白质-多肽对接领域,并评估其性能。利用标准蛋白质-多肽对接数据集(peptiDB)的21个测试例及13个定制测试例进行性能测试,主要对比分析了各算法的成功率和对接精度。结果表明FIPSDock可用于蛋白质-多肽对接,并相对于其他对接算法展示出优异的性能。

关键词: 多肽对接, 蛋白质-多肽间相互作用, 结合位点预测, 对接优化

Abstract: Protein-peptide interaction is prevailing in computational biology, the accurate binding site prediction of protein-peptide is of great importance for understanding signal pathway in vivo, protein functionality and rational drug design. Since computational protein-peptide docking methods can provide sufficient modeling, the performance of them is crucial for experimental accuracy and docking success. In this paper, by reviewing current research progress in protein-peptide docking, it has a comprehensive understanding of protein-peptide docking methods performance, and especially extends the Fully Informed Particle Swarm Dock (FIPSDock) docking method to protein-peptide complexes for test. Experiments are conducted on 21 test cases of standard peptide docking benchmark peptiDB and another custom dataset with 13 cases, to have a comparative comparison of docking accuracy and success rate, the results demonstrate feasibility and novel performance of  FIPSDock for peptide docking.

Key words: peptide docking, protein-peptide interaction, binding site prediction, docking minimization