Identification of candidate regulatory SNPs by integrative analysis for prostate cancer genome data

Genome-wide association studies (GWAS) have identified numerous single nucleotide polymorphisms (SNPs), also known as generic variants, associated with disease susceptibility. Prostate cancer (PCa) is a highly heritable disease. GWAS studies have so far reported more than 70 SNPs that are associated with PCa risk. However, most of these SNPs are located in the noncoding genomic regions that little are known about their functional roles. Here we describe an informatics system that performs an integrative analysis of ChIP-seq, RNA-seq, SNP array and clinical data for identifying candidate regulatory SNPs (rSNPs) that could alter transcription factor (TF) binding sites and neighboring gene regulation. By applying the informatics framework on HOXB13 TF in PCa, we identified 213 rSNPs that include a recently discovered rSNP (rs339331) and identified a novel candidate rSNP (rs1476161) associated with the PCa risk. We confirmed rs1476161 by performing the HOXB13 knockout experiment. The expression level the target gene, AURKB, was decreased by about 2-fold in HOXB13-silencing cells compared to the control cells. This indicates the involvement of HOXB13 in altering AURKB gene expression, suggesting a critical role of rs1476161 in allelespecific gene regulation. Taken together, the results demonstrate the feasibility of our system in searching for candidate rSNPs associated with PCa risk.