Centre for Computational Biology

Welcome

NIH Innovation in Biomedical Information Science and Technology project (2004-2007)

Detecting Relationships among Heterogeneous Genomic Data

Principal investigators

• William S. Noble (PI)
• David Baker (Washington University)
• Nello Cristianini (University of California, Davis)
• Michael Jordan,(University of California at Berkeley)
• Laurent El Ghaoui (University of California at Berkeley)
• Jean-Philippe Vert (Ecole des Mines de Paris).

Presentation

Several objects in computational biology can be characterized by different sorts of data. For example, genes can be represented by a nucleotide or amino-acid sequence, a 2D or 3D structure, a promoter region, expression profiles, localization in the cells, position in the interactome etc... The goal of this project is to develop systematic approaches to integrate such heterogeneous data and learn from them. The targetted learning tasks are the functions and interactions of proteins on a large scale.

The approach are based on recent development of statistical learning theory, kernel methods and optimization theory.

Support

This project is supported by NIH grant R33HG003070-01