BETHESDA, Md.--The US National Institutes of Health awarded $12.8 million to investigators at nine projects that will contribute to the creation of a public pharmacogenetics database. The tool could become available for public use within one year.
The bulk of the money will fund research that will generate the content of the database--information on drug responses in specific patient categories. But three investigators were awarded grants to create and apply bioinformatics tools that will make the data accessible.
Russ Altman of Stanford University was awarded $1.6 million to operate the Stanford Pharmacogenetics Knowledge Base (PharmGKB). NIH will give $421,000 to Yale University medical informatics professor Prakash Nadkarni to design web-based tools for incorporating existing pharmacogenetics knowledge into that database. And Richard Weinshilboum at the Mayo Foundation in Rochester, Minn., will use a $576,000 award to search for variations in genes encoding proteins already known to be important in the body's handling of an array of medicines, hormones, and chemical messengers.
The one-year funding initiative, to which six NIH institutes, including the National Human Genome Research Institute, are contributing, aims to "forge an interactive research network of investigators who will store data in a shared information library freely accessible to the scientific community."
Neither Altman nor Weinshilboum could be reached for comment, but Nadkarni described his role in the project as "software provider."
He said each pharmacogenetics research project will analyze its own experimental data using tools such as statistical analysis software before submitting them for entry into the PharmGKB database.
Nadkarni's task is to design an interactive database and tools according to Altman's specifications to enable researchers to cross reference the their data with reference data housed in other relevant online databases. For instance, he pointed to a database he designed for the SenseLab project, which lets users link from olfactory receptor experimental data entries directly to relevant Medline publications.
For the pharmacogenetics network, Nadkarni has proposed using the National Library of Medicine's Unified Medical Language System to design a "smarter" database search tool that scans text not just for matching phrases, but also for words with similar meanings. He explained, "Simple word indexing doesn't consult a thesaurus, but as it happens, there is a humongous thesaurus of all medical terms and quite a lot of bioinformatics terms in the Unified Medical Language System."
The terms of Nadkarni's grant permit him to copyright his tools, but require that they be released for public use.
Initially, five pharmacogenetics research projects will generate data to be fed into the database. Brigham and Women's Hospital in Boston will study genetic influences on responses to three types of asthma treatments; Georgetown University Medical Center will investigate variable responses to tamoxifen; University of California, San Francisco, will study how drug response is affected by variability in genes that encode membrane transporters, which interact with a third of most commonly used prescription drugs; University of Chicago will study variations in benefits and toxic side effects of certain chemotherapy drugs; and University of California, Los Angeles, will study Mexican-Americans' response to different antidepressant drugs.
Nadkarni noted that while the database will be accessible to multiple user groups, not all will have privileges that enable them to add or delete information, and access will be restricted according to area of study. "When you log in you will see only the studies that you have permission to see. People working in cancer won't even see psychiatry studies," he remarked. No identifying patient information will be entered into the database, Nadkarni added.
NIH also alotted $322,000 to the University of Houston Law Center to study ethical, legal, and social consequences of using pharmacogenetic data.
The institutes are already soliciting applications for a second round of funding and will give preference to applicants with plans for research that will complement the existing studies.