Using the Transcriptome for SNP and Gene Annotation

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): While we are strongly supportive of NIH initiatives in data sharing, we have long believed that it is insufficient to share only the raw data from genomic studies. The massive amounts of data created in today's genomic studies generate even more massive results that merit more careful scrutiny than is generally practical except through the use of sophisticated databases. Thus, we have devoted substantial resources to developing results databases (see, for example, ://www.scandb.org) that we make publicly available. The SCAN database (SNP and Copy number ANnotation) allows users to query results of our transcriptome studies by SNP, by gene and by region, and can be used to annotate SNPs with information on function, including potential eQTL (expression Quantitative Trait Locus) function. Our preliminary studies with SCAN have revealed that SNPs associated with complex traits are more likely to be eQTLs than minor-allele-frequency matched SNPs drawn from high-density SNP genotyping platforms. These results are robust across a wide range of definitions for trait-associated SNPs and eQTLs (p-values ranging from 10-4 to 10-8), and across a broad range of complex trait phenotypes. We now propose to extend the SCAN database to include results of transcriptome association studies being conducted at the University of Chicago on a broad range of human tissues and to continue to develop software tools to maximize the utility of this database. These efforts are informed by our near-complete immersion in studies relating genotype to phenotype (and in developing methods for relating genotype to phenotype) for many different complex traits. Thus, our specific aims are: 1) to extend the SCAN database to serve results of transcriptome studies in liver, brain adipose tissue, and skeletal muscle in addition to the results of the transcriptome studies from GTEx and our studies in lymphoblastoid cell lines that are currently served; 2) to augment the novel software tools we have already developed for use with the SCAN database to use transcriptome association results to facilitate the identification of genetic risk factors for complex traits; and 3) to develop novel approaches for investigating the function of genetic variation with an emphasis on GxG interaction and nQTLs (network QTLs). PUBLIC HEALTH RELEVANCE: We have long been committed to the development of public results databases (see, for example ://www.scandb.org) that permit us to serve the massive results of genomic studies in a way that facilitates further discovery research. Our project will allow users to query results of transcriptome association studies in a variety of human tissues, and to use this information to discover and better characterize genetic risk factors for complex traits.