期刊
JOURNAL OF COMPUTATIONAL CHEMISTRY
卷 38, 期 3, 页码 169-177出版社
WILEY
DOI: 10.1002/jcc.24667
关键词
random forest; docking; scoring function; protein-ligand binding affinity; machine learning
资金
- NIH [R01-GM079223]
- NATIONAL INSTITUTE OF GENERAL MEDICAL SCIENCES [R01GM079223] Funding Source: NIH RePORTER
The development of new protein-ligand scoring functions using machine learning algorithms, such as random forest, has been of significant interest. By efficiently utilizing expanded feature sets and a large set of experimental data, random forest based scoring functions (RFbScore) can achieve better correlations to experimental protein-ligand binding data with known crystal structures; however, more extensive tests indicate that such enhancement in scoring power comes with significant under-performance in docking and screening power tests compared to traditional scoring functions. In this work, to improve scoringdocking- screening powers of protein-ligand docking functions simultaneously, we have introduced a Delta vinaRF parameterization and feature selection framework based on random forest. Our developed scoring function Delta vinaRF20, which employs 20 descriptors in addition to the AutoDock Vina score, can achieve superior performance in all power tests of both CASF-2013 and CASF-2007 benchmarks compared to classical scoring functions. The DvinaRF20 scoring function and its code are freely available on the web at: https://www.nyu.edu/projects/yzhang/DeltaVina. (C) 2016 Wiley Periodicals, Inc.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据