Cluster Based Selection of Candidate Genes in Genome Wide Association Studies.

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): The candidate's overall goal for this mentored career development award is to gain the expertise and training to enable her to become an independent researcher in statistical genetics and bioinformatics, and to act as a facilitator for cross-discipline collaborative research. To achieve this aim, the candidate has compiled a mentoring team consisting of experts in a variety of fields, all of which have a proven track record of successful collaboration. Genome wide association (GWA) studies may result in hundreds, if not thousands, of SNPs showing significant association with the trait(s) under investigation. For many complex traits, such as obesity, there are arguably thousands of genuinely influential polymorphisms with small effects. Hence, one of the challenges that will be faced by investigators will be the selection and prioritization of SNPs from within the lengthy lists of apparently associated SNPs. Building upon research from related high-dimensional biology fields such as microarray based gene expression studies, the candidate proposes to address the issue of candidate gene selection and interpretation of GWA studies. The proposed research introduces the use of cluster analysis in GWA studies, which will group genes together based on their similarity derived from functional annotations such as Gene Ontology and biochemical pathway information. Specifically, the aims of this project are: 1) implement an automated procedure to identify genes in the study and retrieve annotation information, 2) cluster genes based on shared annotations, and 3) evaluate a battery of methods for determining which clusters of genes are associated with the phenotype. By testing the clusters, instead of all the markers in the study, the multiple test correction burden is decreased and power is increased. The identification of genetic contribution to complex diseases such as obesity, heart disease and diabetes will be key in developing an understanding and and [sic] possible treatments for the many health problems associated with these and other diseases. The methods utilized will be packaged as a web-based software which will be made freely available. The components of the mentored career development award will build upon the candidate's background in molecular biology, biochemistry and statistical genetics in order to yield an independent investigator capable of leading future research into the inheritance of human genetic diseases.