PennCNV,18 a free, open-source tool, was used for copy number variation detection. The input data for the program are log R ratio, a normalized measure of the total signal intensity for the two alleles of the SNP, and B-allele frequency, a normalized measure of the allelic intensity ratio of the two alleles. Values of these quantities are derived with the help of control genotype clusters (HapMap samples), using Illumina BeadStudio software. A hidden Markov model is then used to make CNV calls based on the probability of a given copy state at the current marker as well as the probability of observing a copy state change from the previous marker to the current one. A built-in correction model for GC content19 is included. As the samples were genotyped on several different types of Illumina BeadArray chips (HumanHap300, Human-Hap300-duo, HumanHap550, HumanHap610), we analyzed them with a twofold approach: first, using the full complement of markers on the chip and second, using only a subset of markers, present on most of the chip types, to ensure similar resolution of markers covering the
