Development of a methodology to associate alternative polyadenylation with complex traits

Alternative polyadenylation (APA) is one of the post-transcriptional regulatory processes contributing to human transcriptome diversification. The expression of alternative mRNA isoforms with different 3’ untranslated region (UTR) length can affect mRNA metabolism and localization. Similarly to gene expression, alternative polyadenylation quantitative trait loci (apaQTL) were identified as loci containing genetic variants (SNPs, single nucleotide polymorphisms) with effect on relative expression of the short or long isoform. Genome-wide association studies (GWAS) have uncovered tens of thousands of associations between common genetic variants and complex diseases. However, these statistical associations can rarely be interpreted functionally and mechanistically. Transcription-Wide Association Studies (TWAS) associating gene expression to complex traits, highlighted the possibility to use genetically regulated molecular phenotypes to gain insights into GWAS hits causality and biological mechanisms underlying SNP-trait associations. Here we used RNA-seq data from lymphoblastoid cell lines sampled from 373 European individuals to perform the analysis for both alternative polyadenylation and expression, used as control. Using the FUSION tool we combined the imputation of short/long expression ratios (m/M) and publicly available GWAS summary data to detect associations between alternative polyadenylation and quantitative traits or complex diseases. Our pilot study carried on a small database revealed the feasibility of an APAWAS (APA-Wide Association Study). We aim at applying this methodology on a larger and more complete dataset such as GTEx in order to perform a deep analysis of all human tissues and potential associations with numerous complex traits and diseases.