Developing structure-activity relationships (SARs) of molecules is an important approach in facilitating the hits exploration in the early stage of drug discovery.  Although biological activities of millions of compounds are currently available in public molecular libraries such as PubChem, it can be a very challenging task to find meaningful and novel SAR from these resources.  In this presentation, we introduce a bioactivity-centred clustering approach to group 803,268 compounds on the basis of both structural similarity and bioactivity information, which aimed to mine millions of bioactivity data in PubChem to discover useful SARs.  The bioactive small molecules were clustered according to the two-dimensional (2-D) and three-dimensional (3-D) similarities under three scopes: (1) active in a given bioassay, (2) active for a given protein target, and (3) active for a given pathway.  The resulting 18 million clusters were delivered in a way that each cluster contains a group of small molecules both structural and biologically similar, allowing the user to infer a SAR.  The bioactivity clusters pre-computed using publicly available chemical information enable one to quickly navigate, narrow down the compounds of interest.  Each derived bioactive cluster can be useful resource in developing a meaningful SAR or enable one to design or expand compound libraries from the cluster.  It can also help to predict the potential therapeutic effect and pharmacological actions of less-known compounds from those of well-known compounds (i.e., drugs) in the same activity cluster.

PubChem structure-activity relationship (SAR) clusters

246th National Meeting (2013)

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