Hilary Finucane
Massachusetts Institute of Technology
Partitioning heritability by functional category using summary statistics
Recent work has demonstrated that some functional categories of the genome contribute disproportionately to trait heritability. Partitioning heritability is traditionally done using a variance components approach; however this approach is not feasible at large sample sizes, and there are many datasets for which only summary statistics are available. Here, we introduce a new method for partitioning heritability that requires only GWAS summary statistics and LD information from a reference panel. Our method is similar to a multivariate extension of LD Score regression. It is robust to population structure and to multiple causal variants at a locus, and performs well in simulations. We applied the method to summary statistics from large GWASs of ten phenotypes and found many significant enrichments. Conserved Regions showed a large enrichment (8-25X) across phenotypes, while FANTOM5 enhancers showed huge (40-60x) enrichments in autoimmune diseases but non-significant enrichment in other traits. A cell-type specific analysis showed enrichments consistent with current knowledge of disease biology, as well as a clear enrichment of CNS cell types in both BMI and college attendance. This analysis of summary statistics would not be possible with current methods.
