the LD structure, and (c) computes the block-based p-value as Pblock=min(mep(j)me(j)). Lastly, HYST uses the scaled chi-square test to combine the n LD block-based p-values as to obtain the overall p-value for the gene/pathway 16. The LD blocks were inferred 14 based on the 1000 GenomesHaplotypes Phase 3 reference panel ALL (downloaded from ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/20130502/). We opted for the 1000 Genomes as this is the highest resolution genetic map to date, derived based on a powerful design and utilizing the latent genomic information21. Furthermore, the use of this map will facilitate any follow-up analyses in the current GWAS samples as these often include imputed data based on the 1000 Genomes reference panel. Both the gene-based tests and the pathway analyses were conducted genome-wide and utilized the unweighted form of HYST. Variants were assigned to at least one of the ∼24,100 genes (or within a region extended 5 kb at the 5′ and at the 3′ end of the gene) based on the UCSC RefGene (hg19) coordinates. Variants missing LD information were discarded. For the pathway analysis, we considered 1320 canonical pathways collected from the Molecular Signature Database 22 (manually curated pathways originating from Reactome 23, KEGG 24, and BioCarta 25 databases),
