compute_ML_rho: Maximum likelihood of recombination rate

In Cronjaeger/HotBirdHMM: Find Recombination Hoitspots using a Hidden Markov Model

Description

Approximate the maximum likelihood estimate of the recombination rate. This is done in two steps (each of which relies on optimize()): one to find the approximate scale of rho; followed by a step to find the maximum likelihood on that scale.

Usage

compute_ML_rho(Y, M, theta = NULL, magnitude_max = 1)

Arguments

Y	A binary sequence classifying all observed sites as either homozygous (encoded Y[i] == 0) or heterozygous (encoded Y[i] == 1)
M	the number of hidden states to be used in the HMM
theta	The mutation rate to be used (defaults to Watterson's estimator).
magnitude_max	upper magnitude bound on rho, i.e. magnitude_max = k indicates that we firmly believe rho<10^k to be the case; defaults to +1.

Value

a maximum likelihood estimate of the recombination rate between two sites.

Cronjaeger/HotBirdHMM documentation built on May 9, 2019, 6:01 p.m.