Post-imputation graph: Posterior probability vs. Empirical CDF

Question

[rednalf]

This plot is showing the distribution of the posterior probabilities of the most likely alleles for KIR2DS4TOTAL locus, shown as an empirical cumulative distribution function. What does this mean biologically? My knowledge in statistics is unfortunately quite poor...

Note: two curves are plotted here - one for the input dataset and, for comparison, one for OOB (out of bag) imputations from the reference panel. And from what I know, if the KIR loci are imputed well the two distributions should be similar.

Note: the figure has been produced using KIR*IMP method (http://imp.science.unimelb.edu.au/kir/).