Skip to main content
. 2016 Oct 18;10(Suppl 7):103–108. doi: 10.1186/s12919-016-0013-1

Fig. 2.

Fig. 2

Frequency distribution of the weights used for variants in the weighted covariance kernel. Each weight represents the relative proportion of the functional potential − loglikelihood estimation of each variant in the kernel. The large proportion of variants in the first bin have a very small weight, and correspond to variants with a low confidence of having an allele-specific chromatin accessibility effect