Figure 1

Two approaches for prioritizing disease causal genes from whole-genome or exome sequencing data. (a) The probabilistic scoring approach collects relevant information from multiple data sources, and compiles a statistical model that ranks all genes in the genome by their likelihood of being disease causal. (b) The stepwise reduction approach removes variants that are unlikely to be disease causal based on a series of filtering criteria, until a small set of candidate genes is found. The first approach may be more effective and rigorous, yet the second approach may be easier for non-specialists to understand and interpret. GWAS, genome-wide association study; HGMD, Human Gene Mutation Database; I, indel; LSDB, locus-specific database; NS, non-synonymous; OMIM, Online Mendelian Inheritance in Man; SS, splice acceptor or donor site.