Skip to contents

Binomial thinning of the 1 Mb window counts (x; probability frac) with the expected counts scaled by frac and 1 - frac: the depth analogue of split_votes(). Clone discovery uses one half and the merge / split tests the other, so a depth difference that formed two leaves is not tested on the same counts.

Usage

split_win(win, frac = 0.5, seed = 1)

Arguments

win

1 Mb window depth list(windows, x, e)

frac

fraction of the counts in the first half

seed

RNG seed

Value

list(disc, test): window depth lists