An Evaluation of Population Differentiation for Two Anurans (Rana luteiventris and Hyla regilla): The Limits of Genetic InferenceEPA Grant Number: U914713
Title: An Evaluation of Population Differentiation for Two Anurans (Rana luteiventris and Hyla regilla): The Limits of Genetic Inference
Investigators: Call, Douglas R.
Institution: Washington State University
EPA Project Officer: Michaud, Jayne
Project Period: January 1, 1995 through January 1, 1997
Project Amount: $102,000
RFA: STAR Graduate Fellowships (1995) Recipients Lists
Research Category: Academic Fellowships , Ecological Indicators/Assessment/Restoration , Fellowship - Ecology
The main objectives of this research project were to: (1) develop seven microsatellite markers; and (2) study their frequency distributions in eight Rana luteiventris (three loci) and three Hyla regilla (four loci) populations.
Allele sizes appeared to conform to a stepwise mutation model, except 11 percent of allele differences probably resulted from larger mutations. Most loci were in Hardy-Weinberg equilibrium, although there was evidence of null alleles for three R. luteiventris loci. Significant regressions of M on geographic distance (P < 0.001) suggested a neighborhood size of 19 to 274 frogs for R. luteiventris populations. Using these estimates with models for effective population size, the mutation rate was estimated to be near 10-3. New statistics derived specifically for microsatellite data appear to have higher variances than estimators based on an infinite allele model.
A multilocus detection method for measuring microsatellite variation simultaneously at many loci was investigated. This technique reveals a series of alleles within a restriction fragment profile, but potential measurement errors require grouping alleles of similar size into bins. Mean band sharing and heterozygosity were highly correlated (r = -0.99), and the former increased continuously with increasing bin width. Replicate gels were used to calibrate a binning algorithm and significant differentiation between R. luteiventris (n = 5) and H. regilla (n = 2) populations (P 0.008) was found, which was consistent with results from single-locus markers. Contrary to published findings, band-sharing statistics do not exhibit excessive levels of covariance.