Figure 6
From: TagDust2: a generic method to extract reads from sequencing data
Sample to sample correlation can be improved by using unique molecular identifiers. The right panel shows correlation between samples using 15 and 25 PCR cycles without using the UMI sequences. The left panel shows the same data after collapsing all reads mapping to the location and containing the same UMI. TagDust correctly identified PCR artifacts based on their UMI sequences.