tetra: an improved program for population genetic analysis of allotetraploid microsatellite data

Population genetic analysis of allotetraploid microsatellite data has lagged far behind that of diploid data, largely because of an inability to determine allele copy number for partial heterozygotes. tetrasat developed by Markwith et al. (2006) uses an iterative substitution process to account for all probable combinations of allele copy numbers in populations with partial heterozygote samples. However, tetrasat cannot deal with microsatellite data containing more than 15 partial heterozygotes, because of an exponential increase in genotype combinations. tetra can handle the microsatellite data containing infinite partial heterozygotes. In the program tetra, the frequencies of alleles are measured as the probability with which the known alleles occur in unknown allele locations. The Hardy–Weinberg expected heterozygosity and Nei's coefficient of gene differentiation are calculated based on allele frequencies. The mean and standard error of expected heterozygosity are estimated through bootstrap method.