morton-undocumented-neiD: undocumented morton functions - neiD

In andrew-hipp/morton: Tools for manipulation of phylogenetic, genetic, and other biodiversity data

undocumented morton functions - neiD

Description

morton-undocumented functions are currently undocumented

Details

morton is a work in progress. If you are interested in a function that is not yet documented, please contact Andrew Hipp ([e-mail]).

## Unbiased estimate of D based on Nei 1978, Genetics 89, equation (6) ## Arguments: ## genotypes = a data.frame with column 1 = "locus", column 2 = "allele", and columns 3:length(names(genotypes)) = individuals ## n = a matrix with rows = loci, columns = individuals ## nomenclature from Nei: ## p[i], q[i] = frequency of allele i in populations X and Y respectively ## x[i], y[i] = corresponding sample allele frequences ## G[x] = mean(sum(p[i]^2)) over all loci in the genome (for the population) ## G[y] = mean(sum(q[i]^2)) ... ## G[xy] = mean(sum(p[i]*q[i])) ... ## J[x], J[y], J[xy] = corresponding sample means ## D = -log(G[xy] / sqrt(G[x] * G[y])) – this is the biased D ## to makes this unbiased – Dhat – use Ghat in the place of G, calculated as: ## Ghat[x] = mean((2 * n[x] * J[x] - 1) / (2 * n[x] - 1)) over loci studied ## Ghat[y] = mean((2 * n[y] * J[y] - 1) / (2 * n[y] - 1)) ... ## Ghat[xy] = J[xy] ## thus, equation (6) is: Dhat = -log(Ghat[xy] / sqrt(Ghat[x] * Ghat[y])) ## Note that in equation (12), Nei gives an unbiased estimate of single locus genetic distance based on the minimum distance for the kth locus as: ## d[k] = (2 * n[x] * sum(x[i]^2) - 1) / (2 * (2 * n[x] - 1)) + (2 * n[y] * sum(y[i]^2) - 1) / (2 * (2 * n[y] - 1)) - sum(x[i] * y[i]) ## Dhat-m = mean(d[k]) ## Andrew Hipp ([e-mail]), May 2008

andrew-hipp/morton documentation built on April 7, 2024, 12:15 p.m.