Whole Exome Sequencing and CNV Analysis

Peripheral blood samples were obtained and DNA was extracted using the QIAamp DNA Blood Midi Kit (Qiagen, Germany). WES samples were captured with the Agilent Sure Select Target Enrichment kit (V6 58 Mb; Agilent Technologies, United States) and sequenced on the Illumina HiSeq 2500 platform (Glessner et al., 2014 (link); Li et al., 2015 (link)). CNV coordinates were converted to the GRCh37/hg19 build using the UCSC Genome Browser LiftOver tool. CNVs with 50% or larger overlap with telomere, centromere, segmental duplications, or immunoglobulin regions were excluded (Hanemaaijer et al., 2012 (link)). After filtering we screened out rare CNVs by comparing with the Database of Genomic Variants (DGV¹) and Online Mendelian Inheritance in Man (OMIM²).