Computer-Aided Drug Design. The Computer-Aided Drug Design (CADD) Group is a research unit within the Chemical Biology Laboratory (CBL) that employs, analyzes, and develops computer-based methods to aid in the drug discovery, design, and development projects of the CBL and other researchers at the NIH. We split our efforts about evenly between support-type projects and research projects initiated and conducted by CADD staff members. We are implementing many projects, and making available resources developed by the CADD Group, in a Web-based manner. This offers three advantages: (1) it frees all users, including the group members themselves, from platform restraints and the concomitant expenses for specific software/hardware, (2) it makes resources and results immediately available for sharing among all collaborators regardless of their location, and (3) helps, without additional effort, further the mission of the NCI as a publicly funded institution by providing data and services directly to the (scientific) public.

Chemical Identifier Resolver (CIR). CIR works as a resolver for many different chemical structure identifiers (e.g. chemical names, InChI, SMILES etc.) and allows one to convert the given structure identifier into a full structure representation or another structure identifier including references to particular databases in which the corresponding structure or structure identifier occurs. CIR offers a simple to use, programmatic application programming interface (API) based on URLs requested by HTTP. This allows easy linking of CIR and its content to other scientific web services and program packages. CIR currently provides access to 120 million structure records.

Enhanced NCI Database Browser. The Enhanced NCI Database Browser can be used to search the 250,000-compound Open NCI Database. This dataset is the publicly available part of the half-million structure collection assembled by the NCI’s Developmental Therapeutics Program during the program’s 50+ years of screening compounds against cancer and, more recently, AIDS. Visit the CADD Group’s home page or the Enhanced NCI Database Browser service for more information.

Fundamentals of Protein-Ligand Interactions. The non-covalent binding of a drug to the binding site of an enzyme (or other biomacromolecule) is the fundamental process of most drug actions. In spite of a vast body of experimental data available on protein-ligand complexes, mostly obtained by X-ray crystallography, there are still open questions of how this binding process occurs at the atomic and quantitative energetic level. One of the issues is the range of conformational energies one can expect to find for the small-molecule ligand bound to proteins, which we found to be higher than generally assumed. This has led us to broader questions regarding x-ray crystallographic methodologies, such as whether quantum-mechanical refinement (or re-refinement) of protein ligand structures may improve structural quality in various ways.

HIV Integrase. A long-standing interest of our group has been HIV integrase (IN) as a drug development target. This enzyme catalyzes the integration of the viral DNA into the human DNA, which is an essential step in the viral replication cycle. Only a handful of approved drugs so far are based on IN inhibition. We have been utilizing all available experimental results, be they structural, mechanistic, or biochemical, to model and better understand inhibition of IN by small molecules. A recent expansion of these efforts is our work aimed at developing HIV microbicides for the prevention of infection with HIV by topical application such as vaginal gels.

Among our main collaborators are Stephen Hughes and Yves Pommier, NCI; Wolf-Dietrich Ihlenfeldt, Xemistry, Germany; Vladimir Poroikov, Russian Academy of Medical Sciences, Moscow; and Raul Cachau, Leidos, FNLCR.