Drug discoveries may be aided by formation of drug-target database

Mar 27 2006

The latest issue of SCIENCE magazine features a brief about the work of UMBI scientists and their Binding Database (BindingDB) project.

The BindingDB is a public, web-accessible database of measured binding affinities of small, drug-like molecules for proteins known to be drug-targets.

The latest issue of SCIENCE magazine features a brief about the work of UMBI scientists and their Binding Database (BindingDB) project. The BindingDB is a public, web-accessible database of measured binding affinities of small, drug-like molecules for proteins known to be drug-targets. BindingDB supports the discovery of new medications by enhancing the availability and utility of these critical data. BindingDB currently holds about 14,000 measurements, for about 8,000 drug-like molecules and about 60 different drug-targets.

The March 17th issue of SCIENCE Magazine reports:

"How tightly a potential drug attaches to its target determines how well the compound will work and what dose patients will need. Researchers can nab binding affinities for about 14,000 compounds at BindingDB from Mike Gilson of the University of Maryland Biotechnology Institute in Rockville and colleagues. Gleaned from the literature, the data indicate the strength of attraction between the compounds and key proteins, such as the caspase proteins that control cellular suicide. You can also upload files of molecules not in the database to compare them to inhibitors of a particular enzyme."

"BindingDB aims to facilitate drug-discovery, the design of self-assembling systems, and the development of predictive computer models of binding," said Dr. Mike Gilson, a Professor at the Center for Advanced Research in Biotechnology. "The current data collection is already substantial, and we are gathering new binding affinities at a rate of about 800 a month, so this resource will become increasingly useful. It also provides the unique capability of allowing a user to upload their own compound catalog and rank the compounds according to the likelihood that they will bind a protein listed in BindingDB. Recently, we have been working to integrate BindingDB with NIH's PubChem and with the IMEx consortium of molecular interaction databases."

BindingDB collects data from the public literature, and scientists are also encouraged to submit their own data directly via the on-line forms. Although the database currently focuses on the binding of proteins with drug-like molecules, it also has a collection of data from the host-guest and supramolecular chemistry literature, and aims to expand this in the future.

The Binding Database project is based at UMBI's Center for Advanced Research in Biotechnology located in Rockville, Maryland. The project, begun in 1997, was initially supported by the National Science Foundation and the National Institute of Standards and Technology. The project is currently supported by a grant from the National Institute of General Medical Sciences, an institute of the National Institutes of Health.