Mapping of cis-regulatory variants by differential allelic expression analysis identifies candidate risk variants and target genes of 27 breast cancer risk loci

Background Breast cancer (BC) genome-wide association studies (GWAS) have identified hundreds of risk-loci that require novel approaches to reveal the causal variants and target genes within them. As causal variants are most likely regulators of gene expression, we hypothesize that their identification is facilitated by pinpointing the variants with greater regulatory potential within risk-loci. Methods We performed genome-wide differential allelic expression (DAE) analysis using microarrays data from 64 normal breast tissue samples. Then, we mapped the variants associated with DAE (daeQTLs) and intersected these with GWAS data to reveal candidate risk regulatory variants. Finally, we validated our approach by functionally analysing the 5q14.1 breast cancer risk-locus. Results We found widespread gene expression regulation by cis-acting variants in breast tissue, with 80% of coding and non-coding expressed genes displaying DAE (daeGenes). We identified over 23K daeQTLs for 2753 (16%) daeGenes, including at 154 known BC risk-loci. And in 31 of these risk-loci, we found risk-associated variant(s) and daeQTLs in strong linkage disequilibrium suggesting that the risk-causing variants are cis-regulatory, and in 27 risk-loci we propose 37 candidate target genes. As validation, we identified five candidate causal variants at the 5q14.1 risk-locus targeting the ATG10, RPS23, and ATP6AP1L genes, likely via modulation of miRNA binding, alternative transcription, and transcription factor binding. Conclusion Our study shows the power of DAE analysis and daeQTL mapping to identify causal regulatory variants and target genes at BC risk loci, including those with complex regulatory landscapes, and provides a genome-wide resource of variants associated with DAE for future functional studies.