This script calculates various distances between individuals based on allele frequencies. The distances are calculated by scripts in the stats or vegan libraries, with the exception of the pcfixed (percent fixed differences) distance.
gl.dist.ind(x, method = NULL, plot = TRUE, verbose = NULL)-- name of the genlight containing the SNP genotypes [required]
-- Specify distance measure [SNP: Euclidean; P/A: Simple]
-- if TRUE, display a histogram and a boxplot of the genetic distances [TRUE]
-- verbosity: 0, silent or fatal errors; 1, begin and end; 2, progress log ; 3, progress and results summary; 5, full report [default 2 or as specified using gl.set.verbosity]
An object of class 'dist' giving distances between individuals
The distance measure for SNP data can be one of
Euclidean -- Euclidean distance as computed by dist() in stat locus.count -- number of loci for which individuals differ, as implemented by dist.gene() in ape allele.count -- number of allelic differences between two individuals, as implemented by diss.dist() in poppr relatedness -- genetic relatedness between individuals (G matrix), as implemented by A.mat() in rrBLUP
The distance measure for Tag P/A data (binary) can be one of
Simple -- simple matching, both 1 or both 0 = 0; one 1 and the other 0 = 1. Presence and absence equally weighted. Jaccard -- ignores matching 0, both 1 = 0; one 1 and the other 0 = 1. Absences could be for different reasons. Dice -- both 0 = 0; both 1 = 2; one 1 and the other 0 = 1. Absences could be for different reasons. Sometimes called the Czekanowski or Sorensen distance. Phi -- binary analogue of the Pearson Correlation coefficient.
Refer to the documentation in the relevant packages listed above.
# NOT RUN {
gl.dist.pop(testset.gl, method="euclidean")
# }
Run the code above in your browser using DataLab