This script writes to file the SNP genotypes with specimens as entities
(columns) and loci as attributes (rows). Each row has associated locus
metadata. Each column, with header of specimen id, has population in the
first row.
The data coding differs from the DArT 1row format in that 0 = reference
homozygous, 2 = alternate homozygous, 1 = heterozygous, and NA = missing SNP
assignment.