Dr. Skolnick is the author of over 380 publications, has an h-index of 85. Dr. Skolnick's current research interests include drug discovery, computational biology and bioinformatics.

The cost of developing a new drug for complex diseases has been estimated to be about 1.4 billion dollars and takes about 14 years. Moreover, the likelihood of a drug being approved is quite low. For example, the overall failure rate for a drug candidate to pass a Phase 1 trial due to safety failures is 86.2%. Moreover, the failure rates from Phase 1 to 2, Phase 2 to 3 and Phase 3 to final approval are 33.6%, 41.7% and 41.0%, respectively, with lack of efficacy responsible for over half of these failures. To improve drug discovery yield, we describe FASE, FINDSITEcomb2.0 Accelerated Side effect and Efficacy prediction, a composite AI method which can be implemented at the beginning of the drug discovery process that can dramatically reduce the number of ligands that need to be screened and which can accurately predict drug side effects and efficacy from the potential drug’s chemical structure alone. To improve the yield of virtual ligand screening, FASE employs the FINDSITEcomb 2.0 ligand homology modeling approach. The results of extensive experimental validation across a diverse collection of protein families including the development of novel antibiotic hits are presented. Then, a newly developed multi-label machine learning method MEDICASCY that only requires the small molecule’s chemical structure to make drug side effect, efficacy, and probable mode of action target predictions is described. MEDICASCY has comparable or significantly better performance than existing approaches that require extensive experimental information including a priori knowledge of the involved protein/ligand pathways, associated enzymatic activity, and indications that is often unavailable for new drug hits. Thus, MEDICASCY is a unique useful pre-filter to increase the likelihood of new drug approval in clinical trials. A free web service for FASE is available for academic users at http://pwp.gatech.edu/cssb/FASE.