Figure 2.

(A) The number of mismatch positions (SNDs) covered by at least one non-reference base for the eight SE data sets, where the number of exon positions is shaded black. The overlap of SNDs across the four SE data sets aligned with (B) MapSplice and (C) TopHat, showing that these positions are highly dependent on read length. (D) Total number of called variants, where the number of exon positions is shaded black. Although MapSplice has a larger number of SNDs, TopHat has more called variants.