Influences of statistical power on studies of population genetic structure, and empirical population structure analysis of the shortfin mako shark (<i>Isurus oxyrinchus</i>)
Loading...
Date
Authors
Duckett, Drew Jerome Lee
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Studies of population genetic structure are necessary to inform conservation and management action. However, the success of such studies depends on a number of factors, including the type of molecular marker and sampling scheme employed. The influence of these factors on statistical power remains understudied. Therefore, the present study examines the statistical power to detect population differentiation through both simulations and empirical resampling. Simulations across a variety of demographic and sampling scenarios show all marker types are powerful in high structure scenarios. Both microsatellites and single nucleotide polymorphisms (SNPs) remain powerful in low structure scenarios, but SNPs do display slightly higher power than both microsatellites and mitochondrial DNA (mtDNA) when a large number of samples and loci are employed. Empirical resampling of SNP, microsatellite, and mtDNA data from the shortfin mako shark, <i>Isurus oxyrinchus</i>, was conducted to provide a real-world example to compare with the simulations. SNP analyses displayed the most significant population comparisons, but show patterns that may indicate an increased frequency of detecting false positives. Additionally, hybridization gene capture was used to examine the population structure of <i>I. oxyrinchus</i>. The technique produced 791 SNPs, which were analyzed using Fst, Discriminant Analysis of Principle Components (DAPC), k-means clustering, and Bayesian clustering. Although each method indicated the presence of population structure, the lack of biological significance observed supports previous findings that <i>I. oxyrinchus</i> behaves as a single panmictic population.