Fig. 1: Pipeline for identification of missense and loss-of-function variants in the multiple myeloma families.

After identification of the families, DNA isolation from the blood samples and whole genome or exome sequencing, variant calling, filtering, and annotation, we used our in-house developed Familial Cancer Variant Prioritization Pipeline v.2 to identify the most likely cancer predisposition variants for multiple myeloma. All variants with minor allele frequency (MAF) < 0.001 that segregated with the disease in the families were filtered by CADD score >20, which indicates the top 1% of potentially deleterious variants in the human genome. For missense variants, the corresponding genes were screened for their intolerance against functional variants using the NHLBI-ESP6500, ExAc, and local data sets as well the ExAC Z-score. The location of the variants was checked for evolutionary conservation using GERP (>2.0), PhastCons (>0.3), and PhyloP (≥3.0). Ten tools were used to predict the deleteriousness of the variants: Sorting Intolerant from Tolerant (SIFT), Polymorphism Phenotyping version-2 (PolyPhen-2) HDIV (HumDiv), PolyPhen-v2 HVAR (HumVar), Log ratio test (LRT), MutationTaster, Mutation Assessor, Functional Analysis Through Hidden Markov Models (FATHMM), MetaSVM, MetaLR, and Protein Variation Effect Analyzer (PROVEAN). For loss-of-function variants (frameshift and stopgain), pathogenic and neutral variants were predicted using MutPred-LOF with a threshold score of 0.50 at a 5% false-positive rate. Human Splicing Finder was used to evaluate the effect of splice site variants, with a yes/no score.