Visualization and Chemical Interpretation of Multi-Target Structure-Activity Relationships Using SOMPLS

Abstract

In quantitative structure-activity relationships (QSAR), partial least squares (PLS) are of particular interest as a statistical method. Since successful applications of PLS to QSAR data set, PLS has evolved for coping with more demands associated with complex data structures. Especially, PLS variants focusing on visualization and chemical interpretation are highly desirable in modeling multi-target structure-activity relationships. In this paper, we employed the self-organized PLS (SOMPLS) approach to predict multiple inhibitory activities against three serine protease receptors (Thrombin, Trypsin and Factor Xa). Volsurf descriptors were used as chemical descriptors.

Evolution of PLS for Modeling SAR and omics Data

“…The hybrid PLS/SIMCA has been developed for solving classification problems 68. A hyphenated regression technique, coupling the transparency of a bi‐directional Kohonen (BDK) with the modeling power of PLS, SOMPLS, has been proposed 69,70. A genetic algorithm‐based PLS (GAPLS) has also been proposed, enabling variable selection within the framework of PLS 71,72.…”