Investigating the structural effects of cancer mutations using pr | 20836
Journal of Proteomics & Bioinformatics

Journal of Proteomics & Bioinformatics
Open Access

ISSN: 0974-276X

+44 1223 790975

Investigating the structural effects of cancer mutations using protein interfaces

2nd International Conference on Proteomics & Bioinformatics

July 2-4, 2012 Embassy Suites Las Vegas, USA

Marketa Zvelebil, Konstantinos Mitsopoulos and Octavio Espinosa

Accepted Abstracts: J Proteomics Bioinform

Abstract :

Recent analyses on protein crystal structures in the Protein Data Bank Europe have revealed a comprehensive set of protein interfaces. Coupled with the comprehensive mutation data from NGS, this gives the possibility of exploring how mutations have structural effects on proteins and therefore useful insights into how such mutations may contribute to the phenotype of cancer genomes. We investigated the properties of cancer mutations with respect to protein structure focusing on parameters such as protein topological area (interface, surface, buried), amino acids composition and secondary structure with the eventual aim of creating a predictor of mutation severity and annotate protein-protein interaction networks with potential disruptions. We analysed SNP mutations from the 1000 Genomes Project, representing healthy individuals? genomes, and COSMIC, representing somatic mutations in cancer genomes. Cancer mutations are generally more deleterious in terms of physicochemical properties and evolutionary conservation. Several substitution-specific differences between cancer and neutral mutations occur in specific areas and secondary structures. Cancer mutations also show higher propensity for post-translational modifications and bond-formation. Some of the individual observed biases in mutation classes can be explained by investigation of the distribution of the mutation?s neighbouring amino acid residues which suggests that formation of disulphide bonds and loss of salt bridges may contribute to the mechanism of protein function disruption. Using our parameters, we have created a linear model to describe our data to function as a predictor for classifying cancer and neutral mutations. We are also investigating whether neighbouring residue fingerprints exist for individual mutation classes.

Biography :

Marketa Zvelebil was born in Prague, Czechoslovakia but, after the occupation of her country in 1968, her family moved to the USA and then to the Netherlands. In 1979, she came to the UK to finish her education. She obtained her PhD at Birkbeck College, University of London. She has carried out several research positions before, in 1993, joining the Ludwig Institute for Cancer Research (UCL) to set up a Bioinformatics team and became a Reader in Bioinformatics at University College London. In 2007, Dr Zvelebil was appointed Team Leader and Reader of Cancer Informatics at the Breakthrough Research Centre.