San Diego CA, USA, February 27, 2002 -- Structural Bioinformatics, Inc. (SBI), based in San Diego, CA, announced today that a new study presented at the 9th Annual Conference on Retroviruses and Opportunistic Infections unveiled a novel computational method for predicting HIV-1 resistance to protease inhibitor medicines.

Structure-based resistance prediction was developed by SBI in collaboration with scientists at Quest Diagnostics Incorporated (NYSE: DGX). Using sophisticated computational technology SBI scientists created detailed three-dimensional molecular models of highly mutated viral proteins based on genetic information which was obtained from HIV-infected patients. Specifically, the structure of HIV protease, an essential protein for HIV replication, was accurately determined complexed to clinically available protease inhibitors. The binding affinity of each of the seven FDA-approved HIV protease inhibitor medicines was determined by calculating the binding energy of each drug to the active site of the protease.. The scientists were then able to demonstrate with a high degree of confidence that the binding energy directly correlates to drug resistance measured using commercially available cell-based phenotypic resistance assays.

"Rules-based resistance interpretation may not accurately predict the additive or canceling effects of multiple mutations in the protease. This is likely to be especially true for newly introduced medicines where we have limited empirical data," said Peter Heseltine, M.D., Medical Director for Infectious Diseases at Quest Diagnostics' Nichols Institute, based in San Juan Capistrano, Calif. "With this technology, we will be able to more rapidly predict drug resistance with an accuracy approaching that of lengthy cell-based phenotypic assays."

"Viral resistance to currently available medicines remains an intransigent problem in HIV infection, with the majority of new patients expected to become resistant to at least one therapy in the course of treatment," said Dr. Edward T. Maggio, Chairman and Chief Executive Officer of SBI. "In addition, the results are a further validation of SBI’s technology approach in the determination of protein structural information."

SBI’s novel computational method for predicting HIV resistance was developed using the Company’s Variome™ Polymorphism Structure Database. The Variome™ database contains high-quality, 3-D protein structures derived from DNA variants (polymorphisms). Based on this extensive collection of variants, SBI’s researchers are able to create a statistical representation of the range of target protein shapes. With this information, the Company’s researchers are able to predict drug activity, as well as synthesize drug candidates with activity against highly conserved regions of each target’s active binding site.

About Structural Bioinformatics, Inc.

Structural Bioinformatics, Inc. (SBI) is an innovative drug discovery company built on a proprietary proteomics platform. SBI has developed advanced technologies that unify 3-D protein structural information with pharmacogenomics, bioinformatics, and chemistry solutions. SBI’s products and services include Genes to Leads^®, a proprietary in silico technology that provides initial small-molecule leads in less than 120 days; ProMax™, a database of more than 6000 3-D protein structures of atomic resolution; the Variome™ structural pharmacogenomics database; and StructureBank™, a relational management database for storing and retrieving structures. SBI has offices in San Diego, Calif.; Cambridge, Mass.; Washington, D.C., and Hørsholm, Denmark. More information may be found on SBI’s website at http://www.strubix.com