Researchers from Boston University’s Department of Genetics and Genomics announced the release of a public website called “GMED” (Genomic Medicine Database) containing pre-publication results from the first comprehensive genome-wide association study of the Framingham Heart Study Cohort. The website is described in the “Netwatch” section of the Sept. 8th, 2006 issue of Science magazine.


GMED’s goal is to rapidly disseminate links between specific human genetic variants and traits like obesity, hypertension, blood glucose and cholesterol levels so that the findings can be compared with results obtained by other scientists and clinicians that are performing genome-wide association studies in other groups. Collectively these studies will lead to the identification of those differences between individuals that make one person more susceptible to disease than others.


The same researchers reported the identification of the first common obesity-predisposing variant earlier this year in the journal Science. Validation of this association in multiple populations around the world indicates that the database is likely to contain many other important findings. The public website contains over 200 additional associations, including other candidate obesity-predisposing genes.


GMED was designed and built by Dr. Marc Lenburg, Assistant Professor of Genetics and Genomics and Mr. David Ulrich of Gene-Home, Inc. “We realized as we did the Framingham genome scan that giving the community rapid access to the data and the results is very important”, Lenburg said. Dr. Lenburg is an expert in information technology and statistics, but began his career as a biologist.


“GMED also turns out to be a valuable tool to shift through the mountains of information generated by a study this large, said Dr. Alan Herbert, a co-leader of the study in the Genetics and Genomics department. “You want to make the findings available to many people from many different realms of biology, especially those who may have insights into how specific genes might cause disease. The database is very intuitive to use and does require a deep knowledge of UNIX or programming in C++. Dr. Herbert, an expert in human genetics, performed the association analyses presented in the GMED database. “All of our raw data has been returned to NHLBI and the Framingham study and is available to qualified scientists so that other analysis approaches can be implemented”.


“Marc Lenburg has done a fantastic job with GMED. He has the rare combination of skills in biology and information technology needed to do something like this” said Michael Christman, Professor and Chair of BU’s Genetics and Genomics Department and the principal investigator of the genome-wide association study. “A lot of genome-wide association studies are underway, and there is a huge need to replicate and validate reported associations. If the data obtained by each group remains secret, this process will take much longer. We hope that all genome-wide association studies will disseminate their candidate associations using the GMED paradigm. This is an important science policy issue.”


“We are in the process of replicating the findings we are reporting through GMED, but feel like the information could be useful to other scientists now. We are also collaborating with the Framingham Heart Study Investigators to identify associations with a variety of additional traits using the genotypic information generated in our study”, Christman said.