Detecting the number of clusters of individuals using the software structure: a simulation study

There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

Abstract

The identification of genetically homogeneous groups of individuals is a long standing issue in population genetics. A recent Bayesian algorithm implemented in the software STRUCTURE allows the identification of such groups. However, the ability of this algorithm to detect the true number of clusters (K) in a sample of individuals when patterns of dispersal among populations are not homogeneous has not been tested. The goal of this study is to carry out such tests, using various dispersal scenarios from data generated with an individual-based model. We found that in most cases the estimated 'log probability of data' does not provide a correct estimation of the number of clusters, K. However, using an ad hoc statistic DeltaK based on the rate of change in the log probability of data between successive K values, we found that STRUCTURE accurately detects the uppermost hierarchical level of structure for the scenarios we tested. As might be expected, the results are sensitive to the type of genetic marker used (AFLP vs. microsatellite), the number of loci scored, the number of populations sampled, and the number of individuals typed in each sample.

Related collections

Most cited references 23

Record: found
Abstract: not found
Article: not found

Microsatellites, from molecules to populations and back

Philippe Jarne, Pierre J.L. Lagoda (1996)

0 comments Cited 168 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: not found

Genetic diversity and introgression in the Scottish wildcat.

M. Beaumont, E M Barratt, D Gottelli … (2001)

This paper describes a genetic analysis of wild-living cats in Scotland. Samples from 230 wild-living Scottish cats (including 13 museum skins) and 74 house cats from England and Scotland were surveyed for nine microsatellite loci. Pelage characteristics of the wild-living cats were recorded, and the cats were then grouped into five separate categories depending on the degree to which they conformed to the characteristics attributed to Felis silvestris Schreber, 1775. Allele frequency differences between the morphological groups are greater than those among the three house cat samples. Analysis of genetic distances suggests that more of the differences between individuals can be explained by pelage than geographical proximity, and that pelage and geographical location are not confounded. Ordination of the genetic distances suggests two main groups of wild-living cats, with intermediates, and one group is genetically very similar to the house cats, while the other group contains all cats taxonomically identified as wildcat based on morphology. A genetic mixture analysis gives similar results to the ordination, but also suggests that the genotypes of a substantial number of cats in the wildcat group are drawn from a gene pool with genotypes in approximately equilibrium proportions. We argue that this is evidence that these cats do not have very recent domestic ancestry. However, from the morphological data it is highly likely that this gene pool also contains a contribution from earlier introgression of domestic cat genes.

0 comments Cited 111 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: not found

Can assignment tests measure dispersal?

Stephen Sarre, D. Tocher, James Berry (2004)

Individual-based assignment tests are now standard tools in molecular ecology and have several applications, including the study of dispersal. The measurement of natal dispersal is vital to understanding the ecology of many species, yet the accuracy of assignment tests in situations where natal dispersal is common remains untested in the field. We studied a metapopulation of the grand skink, Oligosoma grande, a large territorial lizard from southern New Zealand. Skink populations occur on isolated, regularly spaced rock outcrops and are characterized by frequent interpopulation dispersal. We examined the accuracy of assignment tests at four replicate sites by comparing long-term mark-and-recapture records of natal dispersal with the results of assignment tests based on microsatellite DNA data. Assignment tests correctly identified the natal population of most individuals (65-100%, depending on the method of assignment), even when interpopulation dispersal was common (5-20% dispersers). They also provided similar estimates of the proportions of skinks dispersing to those estimated by the long-term mark-and-recapture data. Fully and partially Bayesian assignment methods were equally accurate but their accuracy depended on the stringency applied, the degree of genetic differentiation between populations, and the number of loci used. In addition, when assignments required high confidence, the method of assignment (fully or partially Bayesian) had a large bearing on the number of individuals that could be assigned. Because assignment tests require significantly less fieldwork than traditional mark-and-recapture approaches (in this study 7 years), they will provide useful dispersal data in many applied and theoretical situations.