Fig. 5

Comparison of counts estimated by mutscan, DiMSum, and Enrich2. For each data set, a representative sample is shown (indicated in the respective figure titles, together with the data set). In general, the counts estimated by the three methods show a high correlation, and deviations are likely explained by differences in aspects such as how the set of allowed or forbidden mutations is specified, how the filtering of low-quality sequences is implemented, and whether mismatches or partial matches are allowed in the specified primer sequence