values of −2, −0.3 and 0.25 for the mean LRRs of potential homozygous deletions, hemizygous deletions and duplications, respectively. For males, X chromosome thresholds of −2 and 0.1 were used for hemizygous deletions and duplications, respectively. For females, X chromosome thresholds of −1.5, −0.1 and 0.6 were used for homozygous deletions, hemizygous deletions and duplications, respectively. Initial CNV calls were then further filtered by their B allele frequency distribution patterns. Autosomal and female chromosome X B allele frequency patterns for duplications and homozygous deletions were required to fit a three-copy mode more closely than a two-copy mode. In addition, paucity of AB alleles in the segment was a requirement for calling the segment a hemizygous deletion. For comparative purposes, two CNVs were considered the same or non-unique if their overlapping region contained at least 80% of the SNPs comprising each CNV. We limited the CNV set for consideration to those spanning ⩾10 consecutive SNPs after conducting a trial of >200 experimentally validated CNVs that showed >99% accuracy for our CNV prediction method using this threshold. In addition, all CNVs meeting quality and experimental design metrics were independently assessed by inspection of each BeadStudio visualization.
