Fig. 1

Comparison of predicted error rates with observed error rates. The solid line represents the theoretical, predicted error rate given a Q score, P = 10^(−Q/10), where Q is the Phred quality score and P is the predicted error rate. The actual error rates for all called Q scores are the mean values calculated from all nucleotide positions within all reads for the three datasets: Burkholderia thailandensis (square), Yersinia aldovae (triangle), and Francisella philomiragia (circle). 95 % confidence limits were used as error bars, however, due to the large amount of data sampled, the error bars are too small to be seen, and are covered by the height of the symbol