Fig. 2From: Machine learning on alignment features for parent-of-origin classification of simulated hybrid RNA-seqSummary of accuracy gains. Bars show the parent-of-origin prediction accuracy on simulated hybrid reads. Blue: Read pair origin predicted by aligner’s choice of one best alignment to the concatenation of parental references. Orange: Origin predicted by a model trained on one portion of those alignments and tested on another portion. In the X-axis labels, Arab = Arabidopsis, Bras = Brassica, and Equu = Equus, while Bt = Bowtie2, Hs = HiSat2, and S = STAR. Bowtie2 was used on RNA references, HiSat2 on DNA references, and STAR on both. The vertical axis ranges from 50% (guessing) to 100% (perfect)Back to article page