Fig. 2

Error rates in different error-correction methods. a Detailed profiling of error rate on every nucleotides. Every dot represents the observed error rate on a certain nucleotide. Blue, green, orange and purple represents transition, transversion, insertion and deletion, respectively. The dashed lines represent the value of local regression. Blue arrows indicate some high rate errors. Red arrows indicate a highly possible real mutation. Two technical replicates are plotted on the same subgraph. b Barplot of medians of different error-correction schemes. The labels, ec1f, ec1r, ec2, ec3f, ec3r, and ec4 represent Scheme 1 forward reads, Scheme 1 reverse reads, Scheme 2, Scheme 3 forward reads, Scheme 3 reverse read, Scheme 4, respectively