PharmaMiner

Chemical compound activity prediction can be significantly improved by defining a Joint Pharmacophore Space (JPS) of chemical compounds, targets, and chemical/biological properties. We define this space to be the spatial arrangement of pharmacophoric points, defined using attributes of chemical compounds such as donors, acceptors, aromatic rings, and hydrophobic cores. Since the underlying geometry of the pharmacophores is responsible for binding between compounds and targets as well as properties of compounds such as Blood Brain Barrier (BBB) permeability, analysis of this joint space enables the discovery of unique pharmacophores and drug candidates. Our PharmaMiner tool defines and analyzes this space for a set of user-defined properties. The tool also annotates subspaces of the JPS with physicochemical and biological properties (such as BBB permeability and toxicity), thus making it possible to define structure-based filters early in the drug discovery process.

Competing ideas for drug discovery rely on the use of fingerprints and physicochemical properties (such as clogP and polar surface area). Our competitive advantage is that we use 3D information to represent chemical compounds and are able to consider multiple targets and properties simultaneously with user interactivity. The PharmaMiner tool is also unique in that it can deconvolute and categorize the actives from cell-based assays, and more broadly to define common geometric patterns of binding across all actives.

PharmaMiner 1.0 is scheduled for release in 2H 2011.