Molecular impact of mutations in monogenic disease and cancer
Univ Of Maryland, College Park, College Park MD
Investigators
Linked publications, trials & patents
Abstract
ABSTRACT Next generation genome scale sequencing of patients is now becoming routine for two classes of disease: rare Mendelian traits and cancer. In favorable cases, these data allow identification of relevant mutations and thus aid diagnosis and therapy. In both classes of disease, the most common type of mutation is missense - single base changes that result in an amino acid substitution in a protein. Uncertainty as to the impact of these mutations on in vivo protein activity has resulted in a very conservative approach to their interpretation in the clinic, so causing many missed opportunities for targeted treatment. The goal of this project is to use a combination of three strategies to make the interpretation of these mutations much more applicable in the clinic. There are already a large number of computational methods that attempt to determine the impact of missense mutations on function, and there is substantial evidence that these have useful accuracy. The primary difficulty is that the accuracy in any particular case is not reliably calibrated. Therefore, our first aim is to use a combination of these methods to develop an approach focused on more reliable estimates for the probability of high impact on protein function (i.e. more confident P values). The second aim is to maximize the utilization of three- dimensional structural information, largely ignored by most computational methods. A large fraction of missense mutations in these classes of disease act by destabilizing protein structure and knowledge of structure allows these to be identified with much higher reliability. Also, structure provides a framework for detailed annotation and comprehension of function. To facilitate the utilization of structure, we will implement a modeling platform that leverages available experimental information to maximize the structural data available for analyzing mutation impact. An important aspect of the platform is incorporation of methods for evaluating the reliability of the structural features relevant to analysis of each mutation. In the third aim we will build specific functional models for each protein of interest, integrating information from current databases, the literature, and community input, so as to provide the richest possible background against which to judge the impact of mutations. Proteopedia, a well established media wiki for proteins, will be used to provide an integrated view of text, data, and structure. A key component of the information resource will be contributions from curators, who will provide annotation and also solicit input from other experts. This aspect of the project builds on experience with other crowdsourcing endeavors, including CASP, CAGI and Proteopedia. There will be three primary outcomes from the project: First, improved reliability for the interpretation of missense mutations. Second, a prototype mutation annotation procedure suitable for use in a clinical setting. Third, the resource will provide information of benefit to a range of other scientists, thus facilitating the analysis of disease related mutations.
View original record on NIH RePORTER →