Figure 4
From: Empirical assessment of sequencing errors for high throughput pyrosequencing data
Goodness of fit assessment. χ2 statistics of maximum likelihood fits per context (insertion gaps on the left, deletion gaps on the right) for the Newbler (a), and WGS assembler data (b). The first pair of bars in each graph correspond to any context (context=*). The remaining pairs of bars correspond to the contexts listed in the usual order A 1, C 1, G 1, T 1, A 2, C 2, G 2, T 2,...