Bioinformatics: High throughput prediction of disease caused by SNPs

Many diseases are caused by single nuclear polymorphisms (SNPs) that cause an amino acid in a protein to be mutated. However, most SNPs do not cause a disease. Today SNPs are readily detected in large scale studies of individual genetic variation. Here, we want to develop methods for analysis of molecular consequences following these SNPs and thereby aid in the identification of what SNPs are most likely to be disease causing.

In addition to mapping the structural effects of the mutations it will be of great importance to use information about the pathway roles played by the involved proteins.  We will apply the new methods to predict structural effects of mutants on biomedically interesting proteins, primarily on proteins associated with cancer development and on medically important membrane proteins.