A biologically informed method for detecting associations with rare variants
Moore, Carrie Colleen Buchanan
Many recent studies have identified rare variants that contribute to common, complex disease. It is believed that rare variants likely have a larger effect size (compared to GWAS findings) and can act alone, in concert with other rare variants, or together with common variants. Multiple rare variants can potentially account for a portion of missing heritability in a given trait; therefore, binning or burden testing, may better account for genetic heterogeneity. BioBin, an innovative collapsing method developed in the Ritchie lab, utilizes a flexible repository of data assembled from multiple public databases. The novelty of BioBin lies in access to comprehensive knowledge-guided multi-level binning. BioBin can apply multiple levels of burden testing, including: functional regions, evolutionary conserved regions, genes, and/or pathways. BioBin does not include a specific statistical association test, since the application of statistical testing is dependent on data type and analysis in question. Therefore, the user has the flexibility to apply tests appropriately without constraint. BioBin has been tested in the context of extensive simulation studies, compared with multiple published statistical methods, and applied to the NHLBI GO Exome Sequencing Project for Cystic Fibrosis. BioBin is a very useful and flexible tool to analyze sequence data and can uncover novel associations with complex disease.