Support vector regression scoring of receptor-ligand complexes for rank-ordering and virtual screening of chemical libraries.
Support vector regression scoring of receptor-ligand complexes for rank-ordering and virtual screening of chemical libraries.
复制标题
DOI：
10.1021/ci200078f
复制
发表时间：
2011-09-26
影响因子：
5.6
通讯作者：
Meroueh SO
中科院分区：
文献类型：
--
作者：
Li L;Wang B;Meroueh SO
社区结构活动资源 (CSAR) 数据集用于在回归模式 (SVR) 下开发和测试基于支持向量机的评分函数。推导了两个评分函数（SVR-KB 和 SVR-EP），目的是重现两个 CSAR 数据集中提供的实验结合亲和力的趋势。用于训练SVR-KB的特征是基于知识的成对势，而SVR-EP则基于物理化学性质。将 SVR-KB 和 SVR-EP 与其他七种广泛使用的评分函数进行了比较，包括 Glide、X-score、GoldScore、ChemScore、Vina、Dock 和 PMF。结果表明，使用从 PDBbind 数据集的三维复合体获得的特征进行训练的 SVR-KB 优于所有其他评分函数，包括最佳性能 X 得分，使用三个相关系数（即 Pearson、Spearman 和 Kendall）接近 0.1。有趣的是，排名排序中的更高表现并没有转化为使用有用诱饵目录 (DUD) 的 40 个目标进行评估的虚拟筛选的更大丰富性。为了解决这种情况，我们通过遵循我们之前用于导出 SVM-SP 的目标特定定制策略，开发了 SVR-KB (SVR-KBD) 的变体。 SVR-KBD 表现出比所有其他测试的评分函数更高的富集度，并且在性能上与我们之前导出的评分函数 SVM-SP 相当。
The Community Structure-Activity Resource (CSAR) datasets are used develop and test a Support Vector Machine-based scoring function in regression mode (SVR). Two scoring functions (SVR-KB and SVR-EP) are derived with the objective of reproducing the trend of the experimental binding affinities provided within the two CSAR datasets. The features used to train SVR-KB are knowledge-based pairwise potentials, while SVR-EP is based on physico-chemical properties. SVR-KB and SVR-EP were compared to seven other widely-used scoring functions, including Glide, X-score, GoldScore, ChemScore, Vina, Dock and PMF. Results showed that SVR-KB trained with features obtained from three-dimensional complexes of the PDBbind dataset outperformed all other scoring functions including best performing X-score, by nearly 0.1 using three correlation coefficients, namely Pearson, Spearman and Kendall. It was interesting that higher performance in rank-ordering did not translate into greater enrichment in virtual screening assessed using the 40 targets of the Directory of Useful Decoys (DUD). To remedy this situation, a variant of SVR-KB (SVR-KBD) was developed by following a target-specific tailoring strategy that we had previously employed to derive SVM-SP. SVR-KBD showed much higher enrichment outperforming all other scoring functions tested, and was comparable in performance to our previously-derived scoring function SVM-SP.