Identifying Activity Cliff Generators of PPAR Ligands Using SAS Maps

Structure-activity relationships (SAR) of compound databases play a key role in hit identification and lead optimization. In particular, activity cliffs, defined as a pair of structurally similar molecules that present large changes in potency, provide valuable SAR information. Herein, we introduce the concept of activity cliff generator, defined as a molecular structure that has a high probability to form activity cliffs with molecules tested in the same biological assay. To illustrate this concept, we discuss a case study where Structure-Activity Similarity maps were used to systematically identify and analyze activity cliff generators present in a dataset of 168 compounds tested against three peroxisome-proliferator-activated receptor (PPAR) subtypes. Single-target and dual-target activity cliff generators for PPARα and δ were identified. In addition, docking calculations of compounds that were classified as cliff generators helped to suggest a hot spot in the target protein responsible of activity cliffs and to analyze its implication in ligand-enzyme interaction.