While genome-wide association studies (GWAS) have identified numerous cancer susceptibility alleles, neither the mechanism underlying their action nor the genes they influence are understood. Only by understanding how these SNPs influence biological function can they inform our knowledge of cancer etiology and lead to chemoprevention strategies aimed at mimicking the non-risk (protective) alleles. The objective of this proposal is to develop a computational approach to predict the functional variants responsible for these associations. The central hypothesis is that these functional variants will alter transcription factor binding sites in promoter and enhancer elements, resulting in misregulation of nearby gene(s). The rationale is that by identifying the functional regulatory SNPs at these cancer susceptibility loci, the upstream factors that affect genes whose misexpression influences cancer susceptibility will be identified, potentially opening the door to new methods for cancer chemoprevention. The experimental team's extensive experience with integrating genomic data to analyze genome-wide SNP association studies and ongoing experimental research on regulatory SNPs in cancer is strong preparation for undertaking the proposed studies. The central hypothesis will be tested through completion of the following specific aims: 1. Develop an algorithm to identify regulatory SNPs and the transcription factor whose binding they alter. This algorithm will identify SNPs for which only one allele is predicted to bind a transcription factor and combine this with information on evolutionarily conserved regions and epigenetic marks of transcriptional regulation. 2. Identify putative regulatory SNPs responsible for observed SNP associations with cancer susceptibility. Regulatory SNPs in linkage disequilibrium with known cancer risk loci will be identified and tested for association with cancer through statistical imputation in existin genome-wide SNP datasets. 3. Determine the relationship between regulatory SNPs associated with ovarian cancer risk and expression levels of nearby genes. Regulatory SNPs at ovarian cancer risk loci will be tested for association with expression levels of nearby genes in tumor tissue using SNP and gene expression data from The Cancer Genome Atlas. By identifying regulatory cancer risk variants and, for ovarian cancer, the genes they regulate will enable future biological inquiry and potentially the development of chemopreventative agents for cancer.
Numerous hereditary genetic changes that influence who will get cancer have been identified. By understanding the biology by which these changes alter cancer risk, it may be possible to design therapies to mimic those genetic variations that make individuals less likely to develop cancer, thereby helping to prevent cancer from initially occurring.