Brown J B, Urata Takashi, Tamura Takeyuki, Arai Midori A, Kawabata Takeo, Akutsu Tatsuya
Institute for Chemical Research, Kyoto University, Uji, Kyoto, Japan.
J Bioinform Comput Biol. 2010 Dec;8 Suppl 1:63-81. doi: 10.1142/s0219720010005117.
High accuracy is paramount when predicting biochemical characteristics using Quantitative Structural-Property Relationships (QSPRs). Although existing graph-theoretic kernel methods combined with machine learning techniques are efficient for QSPR model construction, they cannot distinguish topologically identical chiral compounds which often exhibit different biological characteristics. In this paper, we propose a new method that extends the recently developed tree pattern graph kernel to accommodate stereoisomers. We show that Support Vector Regression (SVR) with a chiral graph kernel is useful for target property prediction by demonstrating its application to a set of human vitamin D receptor ligands currently under consideration for their potential anti-cancer effects.
在使用定量结构-性质关系（QSPR）预测生化特性时，高精度至关重要。尽管现有的结合机器学习技术的图论核方法在构建QSPR模型方面很有效，但它们无法区分拓扑结构相同但往往表现出不同生物学特性的手性化合物。在本文中，我们提出了一种新方法，该方法扩展了最近开发的树模式图核以适应立体异构体。通过将其应用于一组目前正在考虑其潜在抗癌作用的人类维生素D受体配体，我们表明具有手性图核的支持向量回归（SVR）可用于目标性质预测。
