Exploration of structure-activity relationship determinants in analogue series.

A computational methodology is introduced to systematically organize compound analogue series according to substitution sites and identify combinations of sites that determine structure-activity relationships (SARs) and make large contributions to SAR discontinuity. These sites are prime targets for further chemical modification. The approach involves the analysis of substitution patterns in "combinatorial analogue graphs" (CAG) and the application of an SAR analysis function to evaluate contributions of variable R-groups. It is applicable to analogue series spanning different potency ranges, for example, analogues taken from lead optimization programs or screening data sets (where potency differences might be subtle). In addition to determining key substitution patterns that cause significant SAR discontinuity, CAG analysis also identifies "SAR holes", i.e., nonexplored combinations of substitution sites, and SAR regions that are under-sampled in analogue series.