De novo design algorithms are useful for drug discovery, but can suffer from challenging obstacles including inadvertent assembly of un-physical molecules, combinatorial explosion in chemical space, and poor convergence. We have implemented a novel de novo drug design method integrated into the infrastructure of the virtual screening program DOCK6. Candidate molecules are constructed from fragments of pre-existing molecules that were previously disassembled at rotatable bonds. Molecular growth can be guided by the standard DOCK energy function or by the new footprint-based scoring function recently developed in our laboratory which employs a per-residue decomposition of protein-ligand interactions. Chemical information including 2D fingerprints, 3D pharmacophores, and volume overlap are being evaluated to improve diversity during growth, while limiting combinatorial explosion. These cheminformatics methods can also contribute to the scoring function, thereby biasing growth toward a target reference structure. Application to development of virus-host fusion inhibitors targeting HIV gp41 will be discussed.

Development of a de novo design platform for DOCK: Application to HIV gp41

244th National Meeting (2012)

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