Activity landscape modeling methods are useful tools to analyze structure-activity relationships (SAR). Among them, structure-activity similarity (SAS) maps are important representations to aid the SAR characterization of molecular databases. SAS maps are constructed by plotting pairwise comparisons of the structural similarity against the absolute value of the activity difference (or activity similarity). These maps can be employed to detect activity cliffs (compounds where small structural changes lead to large activity differences) and activity cliff generators (molecules that form an increased number of activity cliffs compared with other molecules in the dataset). In addition, pairs of compounds in the smooth regions of the SAR and scaffold hops are also easily identified in these maps. Herein we describe the activity cliff analysis of three molecular databases employing SAS maps that integrate chemotype information. The results reveal different tendencies in local landscapes for subsets of molecules classified in the most frequent chemotypes. Also, quantitative characterization of activity cliffs for compound pairs that fall in each chemotype class was carried out employing the Activity Cliff Enrichment Factor (ACEF), a new index proposed in this work. Finally, chemotype classes enriched in activity cliffs are discussed and analysed employing the activity cliff generator concept. This approach helps to analyse the behaviour of each chemotype in a given activity landscapes and to identify scaffolds with high probability to form activity cliffs.

Activity cliffs on chemotype-based activity landscapes

248th National Meeting (2014)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.