Skip to main content
. 2012 Jan 28;28(7):921–928. doi: 10.1093/bioinformatics/bts055

Fig. 4.

Fig. 4.

The KL divergence compares the frequency of k-mers (here, for k=1 and k=4) surrounding the starts of aligned reads to the frequencies expected under the assumption of uniform sampling from within exons. A large divergence indicates significant bias. Plotted here is the divergence from unadjusted read counts as well as after adjusting read counts using each method.