Figure 2

Single molecule tag-identified duplicates represent PCR duplicates. (A) Boxplot of target duplicate rate as a function of DNA input. DNA samples from two tumors were diluted and single molecule tag (SMT)-libraries were prepared. Duplicate rate, as identified by SMT sequence, was higher in the lower input DNA samples, consistent with lower starting complexity. (B) Smoothed scatterplot of the relationship between target duplicate rate, adjusted for sample-specific effects and GC content of insert and primers, and depth of coverage (shown log scale). (C) Barplot of the number of times that independent SMTs (one SMT per target per sample) were seen across targets and samples (black) compared to expectations by Poisson sampling (grey). (D) Motif identified using the 54 SMT sequences observed 10 or more times across all 14 12-mer germline samples and targets. (E) Agreement of allele calls within duplicate clusters for 12-mer and 8-mer SMT sequences. Percent allele call agreement is the percentage of duplicate clusters where all allele calls at the SNP of interest were consistent.