Scaffold analysis of GPCR ligand datasets from literature.

摘要

Maximum Common Substructure (MCS)-based clustering is a highly intuitive method to categorize chemical datasets by common chemical motifs that exist within a chemical dataset.Because molecules are subdivided and grouped by chemical substructure or scaffolds,small changes in molecular structure can be easily identified,thereby facilitating interpretation of large datasets in a chemically intuitive manner.However,MCS-based clustering methods are computationally more demanding than classical clustering techniques that exploit chemical fingerprints or molecular keys.In this study the benefits of conducting MCS-based clustering analysis versus classical clustering methods is addressed.