Extended Data Fig. 6: ARG rescaling from original ARG (a) to a new ARG (b).
From: Robust and accurate Bayesian inference of genome-wide genealogies for hundreds of genomes
For simplicity we only show a single tree. We partition the ARG and count the number of mutations in each interval, and re-scale the interval length so that the expected number of mutations match the observed mutation counts. For example, the highest interval has more mutations mapped to it than the lowest interval, so it is widened and the latter is narrowed.
