Fig. 3

Evaluating the alignment accuracy and consistency of X-Mapper in sequencing samples of various complexities. Penalties shown are example penalties for a single read. a Alignment accuracy was evaluated by comparing the alignment results of the same reads to the same reference genome under the same penalty (alignment score) settings across different aligners. Alignments with the lowest penalty (higher alignment score) were considered optimal, while all other alignments with higher penalties (lower alignment scores) were declared suboptimal. b Alignment consistency was evaluated by comparing the alignment results of the same reads by the same aligner in one reference (Assembly A1) versus a complex reference (multiple genomes). Alignments were considered consistent if (1) the optimal alignments to a simple reference were also reported when aligned to a complex reference, or (2) a complex reference yielded better alignments (with lower penalties or higher alignment scores) than the simple reference, primarily due to assembly errors. The alignment consistency was then compared across aligners