First author: U.S. Department of Agriculture-Agricultural Research Service (USDA-ARS), 24106 N. Bunn Rd., Prosser, WA 99350; second author: USDA-ARS, Department of Botany and Plant Pathology, 915 West State Street, Purdue University, West Lafayette, IN 47907; and third and fourth authors: Department of Plant Pathology, 334 Plant Science Building, Cornell University, Ithaca, NY 14853
Go to article:
Accepted for publication 12 January 2003.
Estimation of genotypic diversity is an important component of the analysis of the genetic structure of plant pathogen and microbial populations. Estimates of genotypic diversity are a function of both the number of genotypes observed in a sample (genotype richness) and the evenness of distribution of genotypes within the sample. Currently used measures of genotypic diversity have inherent problems that could lead to incorrect conclusions, particularly when diversity is low or sample sizes differ. The number of genotypes observed in a sample depends on the technique used to assay for genetic variation; each technique will affect the maximum number of genotypes that can be detected. We developed an approach to analysis of genotypic diversity in plant pathology that makes specific reference to the techniques used for identifying genotypes. Preferably, populations that are being compared should be very similar in sample size. In this case, the number of genotypes observed can be used directly for comparing richness. In most cases, sample sizes differ and use of the rarefaction method to calculate richness is more appropriate. In all cases, scaling either Stoddart and Taylor's G or Shannon and Wiener's H' by sample size should be avoided. Under those circumstances where it might be important to distinguish whether richness or evenness contribute more to diversity, a bootstrapping approach, where confidence intervals are calculated for indices of diversity and evenness, is recommended.
The American Phytopathological Society, 2003