Figure 4.

MDiGS copy number analysis. (A) Median absolute pairwise difference (MAPD) scores were calculated as the log2 absolute difference between adjacent target regions for 10 samples enriched by exome capture and MDiGS to compare region-to-region variability of similar target regions. MDiGS MAPD scores were also calculated for 500 bp adjacent windows to demonstrate reduced region-to-region variability achieved by using long contiguous BAC baits. Large duplications spanning an entire BAC can be detected in (B) small sample sizes using Method 1, which compares individual test samples to normal controls or (C) large sample sizes using Method 2. Simulated data demonstrate how individual test samples with large CNVs can be correctly identified within a multiplexed pool of 96 test samples with unknown CNV state provided that less than 36 samples contain similar CNVs.