Determining the functional impact of non-coding disease associated SNPs identified by genome-wide association studies (GWAS) is challenging. Many of these SNPs are likely to be regulatory SNPs (rSNPs): variations which affect the ability of a transcription factor (TF) to bind to DNA. However, experimental procedures for identifying rSNPs are expensive and labour intensive. Therefore, in silico methods are required for rSNP prediction. By scoring two alleles with a TF position weight-matrix (PWM), and observing the change in PWM score, it can be determined which SNPs are likely rSNPs. Predicting in this manner, however, yields large numbers of false positive predictions. In addition, no method exists that determines the statistical significance of a nucleotide variation on a PWM score.

We have designed an algorithm for in silico regulatory SNP detection called is-rSNP [1]. We employ novel convolution methods to determine the complete distributions of PWM scores and ratios between allele scores, facilitating assignment of statistical significance to rSNP effects. We tested our method on 41 experimentally verified rSNPs, correctly predicting the disrupted TF in 28 cases. We also analysed 146 disease associated SNPs from The Catalog of Published Genome-Wide Association Studies [2] with no known functional impact in an attempt to identify candidate rSNPs. We predicted 11 SNPs to be significantly disrupting the binding of a TF. Of these 11, 8 had previous evidence of the TF being associated with the disease in the literature. In addition, genes associated with each of the diseases were enriched for binding sites of the predicted disrupted TF in 5 cases. These results demonstrate that is-rSNP is suitable for high-throughput screening of SNPs for candidate rSNPs. This is a useful and important tool in the interpretation of GWAS.