Figure 3From: Empirical assessment of sequencing errors for high throughput pyrosequencing dataGap lengths distribution. Histograms of gap lengths with any context (insertion gaps on the left, deletion gaps on the right) for the Newbler (a), and Celera WGS assembler (b). Also shown are the best fit approximations of these histograms using the geometric (green line) and Zipfian (orange line) models.Back to article page