Special seminar on February 1-st at 11 a.m., 
Bldg 38A, 8-th floor conference room


         Bioinformatics Approaches to the Molecular Basis of Disease

                              Merridee Wouters
                Victor Chang Cardiac Research Institute
                               Sydney, Australia

Redox-active Disulfides Thiol-based redox regulation is likely to be
important in pathological conditions involving abnormal redox states
such as cardiovascular failure, neurodegenerative diseases and ageing.
The ability to distinguish between structural and redox-active
disulfides is important for elucidating protein function.
Experimentally, the two types of disulfide can be distinguished by their
redox potentials. Disulfide redox potentials measured in thiol-disulfide
oxidoreductases range from -120mV to -270mV. For disulfides serving
structural purposes, the redox potential can be as low as -470mV.
However individual measurements of this kind are difficult and time
consuming. Computational approaches that can identify and characterize
redox active disulfides will contribute significantly to our
understanding of disulfide redox-activity. Several protein
structure-based computational techniques we have developed for
identifying redox-active disulfides are described. These include the
Forbidden Disulfide technique which identifies disulfides that disobey
known rules of stereochemistry in protein structures such as vicinal
disulfides between adjacent cysteines in the polypeptide chain.

Candidate Gene Prediction GeneTrepid is a computational system to
prioritize genes associated with specific diseases or traits within
genomic loci. The system adopts two novel algorithms: Common Module
Profiling (CMP) and Common Pathway Scanning (CPS). CMP is based on the
hypothesis that genes of similar function will lead to the same
phenotype and identifies likely candidates using a domain-dependent
sequence similarity approach. The CPS method assumes that common
phenotypes are associated with proteins that participate in the same
complex or pathway and applies network data derived from protein-protein
interaction and pathway databases to identify relationships between
genes. Both CMP and CPS use two forms of input data: known genes or
multiple loci. The system has been tested for its ability to predict
disease genes on a test set of 29 diseases with 3 or more known disease
genes. When using known disease genes as input, our combined methods
have a sensitivity of 0.52 and a specificity of 0.97 and reduce the
candidate list by 13-fold. The GeneTrepid webserver is available to
identify genes associated with particular diseases and traits in
user-specified intervals.