Genomic DNA was extracted from peripheral blood leukocytes according to standard procedures, and then exome libraries (Agilent SureSelectXT Reagent Kit; Agilent Technologies) were sequenced on an Illumina HiSeq 2500 at the Genomic Technologies Facility in Lausanne, Switzerland. Bioinformatic analyses were performed as described before.5 (link) Briefly, raw reads were mapped to the human reference genome (hg19/GRCh37) using the Novoalign software (V3.08.00, Novocraft Technologies). Next, Picard (version 2.14.0-SNAPSHOT) was used to remove duplicate reads and GATK was used (version 3.8)6 (link) to perform ‘base quality score recalibration’ on both single nucleotide variants and insertion-deletions. A VCF file with the variants was generated by HaplotypeCaller. Then, DNA variants were filtered based on quality, frequency in ExAC, gnomAD,7 (link) 1000 Genomes,8 (link) ESP (NHLBI Exome Variant Server, http://evs.gs.washington.edu/EVS), GME (GME Variome http://igm.ucsd.edu/gme/index.php) and ABraOM,9 (link) and on predicted impact on protein sequence and mRNA splicing. Finally, they were annotated according to a specific in-house pipeline.5 (link)