Khachik Smbatyan

This paper evaluates autonomous LLM agents in molecular prediction and optimization benchmarks, showing they can develop methods that match or exceed human-developed approaches in several tasks.

This paper introduces DODock and DOScore, hybrid ML/physics frameworks for docking and scoring that improve accuracy while generalizing better to novel molecules, showing significant improvements in hit rates for virtual screening.

The paper introduces SDDF, a platform using volunteer computing and active learning to efficiently generate molecular conformation datasets for drug discovery.