Complete summary of the results after applying BANNs, SuSiE [46], and RSS [26] to high-density lipoprotein (HDL) and low-density lipoprotein (LDL) cholesterol in both individuals assayed within the Framingham Heart Study and ten thousand randomly sampled individuals of European ancestry from the UK Biobank.

The first page compares the overlap of significant SNPs and SNP-sets found by each method according to the “median probability model” threshold [57] (i.e., PIP ≥ 0.5) in the Framingham Heart Study. The second page lists how many SNPs and SNP-sets were replicated for each method when analyzing the independent dataset from the UK Biobank. Results are based on defining gene boundaries in two ways: (a) we use the UCSC gene boundary definitions directly, and (b) we augment the gene boundaries by adding SNPs within a ±500 kilobase (kb) buffer to account for possible regulatory elements.
(XLSX)