Drug Discovery Institute

Structure Based Drug Discovery

Our Structure-Based Drug Discovery (SBDD) capability supports target characterization, small-molecule hit identification, and hit-to-lead development by combining structural information with computational tools in the areas of molecular modeling and simulation and cheminformatics. We also develop software and optimize computational protocols for various projects.

Our Approach

We use a computational approach to understand target structures and interactions with ligands. This is an efficient, economical way to identify compounds that can be used to generate and optimize lead small-molecule drugs. Our objective is to reduce the time and cost for Mount Sinai researchers to discover compounds that modulate the function of specific target proteins. At the same time, we provide detailed structural information about protein-ligand interactions, which can be used to optimize these ligands.

We provide access to databases, state-of-the-art software, and expert support for protein sequence and structure analysis, protein structure modeling, characterization of protein-ligand interactions, virtual screening, and cheminformatics.

Our computational services allow screening of Icahn School of Medicine at Mount Sinai's small-molecule libraries in the absence of a high-throughput assay, and also provide computational screening of larger libraries of millions of commercially available compounds. We work closely with other groups at the Drug Discovery Institute, especially Medicinal Chemistry. These interactions facilitate the experimental validation and optimization of small-molecule compounds identified in computational screens.


In SBDD, we use the following tools to screen molecules:

  • Protein structure modeling (homology modeling, protein-protein docking, etc.)
  • Molecular dynamics simulations
  • Binding site identification and characterization
  • Modeling of protein-ligand complex structures
  • In silico screening of compound libraries (target and ligand-based)
  • Virtual small-molecule libraries containing 20+ million purchasable compounds
  • Bioactivity databases
  • Cheminformatics support for compound selection (e.g. drug-likeness)
  • Cheminformatics for compound fragmentation, comparison, clustering, etc.