Fig. 3
From: Theoretical characterisation of strand cross-correlation in ChIP-seq
Distributions of parameters and theoretical prediction result for real data. a-d Distributions of (estimated) parameters M, \(\hat {n}\), \(\hat {\alpha }\) and \(\hat {w}\). Note that the y-axis for (C) is logarithmic. e Distribution of “degree of saturation” (Mα/2nw, see Eq. (16)). Where the degree of saturation exceeded one (red area), such samples must be treated as saturated cases. f Log-log plot for maximum NCC coefficient comparison between values calculated with PyMaSC (x-axis) and estimated from our model and peak calling analysis (y-axis)