Fig. 3

Identification and characterization of potential pathogenic Lynch syndrome variants. a Analysis workflow: 212 individuals with MSI-H classification were dichotomized based on the presence of germline:somatic mutation of a L-MMR gene. Individuals carrying germline:somatic mutations were further subdivided by allele frequency of the candidate germline variant in ExAC. Pink boxes indicate the use of somatic data, and blue boxes integrate somatic and germline data. Numbers in parentheses refer to number of individuals that fulfill the box criteria. Individuals that carry bi-allelic alterations are labeled according to ClinVar significance of the germline variant. VUS variant of unknown significance. b, c Somatic MSI burden (b) and age of diagnosis (c) of individuals who carry germline:somatic mutations in a MMR gene. Individuals were grouped by MMR gene mutation type: None, no alteration; Germ, germline LOF variants only, Som, somatic LOF mutations only; Bi-Miss, bi-allelic alteration including a missense mutation; and Bi-LOF, bi-allelic alteration via dual LOF mutations. Age was converted to a Z-score using the mean and standard deviation age of diagnosis for each cancer type. **p < 0.001, *p < 0.01; p values were determined using a linear model to predict somatic MSI burden while accounting for cancer type