Postdoctoral Research Fellowship in Interdisciplinary Informatics for FY 2003

Goldberg Debra S, Boston MA

Investigators

Abstract

Postdoctoral Research Fellowships in Interdisciplinary Informatics are sponsored jointly by the Directorates for Mathematical and Physical Sciences (MPS) and Biological Sciences (BIO) to encourage research and training that cross the traditional disciplinary boundaries between them. These fellowships provide opportunities for interdisciplinary research and educational activities in biology and informatics to a wide range of recent doctoral recipients (biologists, chemists, physicists, mathematicians, statisticians, computer scientists, and others). It is expected that the Fellows trained through these fellowships will play an important role in training the future workforce. Postdoctoral research and training in informatics will permit junior scientists trained in biology, mathematical, chemical, and physical sciences to play key roles in developing new quantitative tools and methods that will advance informatics in biology and other fields. The research and training plan is entitled "Protein function prediction from error-prone genomic data using confidence weights and network topology." High-throughput experimental methods are both expensive and error-prone. Still, the underlying biology generates patterns in the global network topology. This study integrates diverse functional genomic data using probabilistic edge weights based on experimental data and shows how each observation fits the expected local graph topology to infer protein function.