Inferring introduction routes of invasive species using approximate Bayesian computation on microsatellite data

T Guillemaud, MA Beaumont, M Ciosi, JM Cornuet, A Estoup

Research output: Contribution to journalArticle (Academic Journal)

115 Citations (Scopus)

Abstract

Determining the routes of introduction provides not only information about the history of an invasion process, but also information about the origin and construction of the genetic composition of the invading population. It remains difficult, however, to infer introduction routes from molecular data because of a lack of appropriate methods. We evaluate here the use of an approximate Bayesian computation (ABC) method for estimating the probabilities of introduction routes of invasive populations based on microsatellite data. We considered the crucial case of a single source population from which two invasive populations originated either serially from a single introduction event or from two independent introduction events. Using simulated datasets, we found that the method gave correct inferences and was robust to many erroneous beliefs. The method was also more efficient than traditional methods based on raw values of statistics such as assignment likelihood or pairwise F(ST). We illustrate some of the features of our ABC method, using real microsatellite datasets obtained for invasive populations of the western corn rootworm, Diabrotica virgifera virgifera. Most computations were performed with the DIYABC program (http://www1.montpellier.inra.fr/CBGP/diyabc/).
Translated title of the contributionInferring introduction routes of invasive species using approximate Bayesian computation on microsatellite data
Original languageEnglish
Pages (from-to)88 - 99
Number of pages12
JournalHeredity
Volume104(1)
DOIs
Publication statusPublished - Jan 2010

Bibliographical note

Publisher: Nature Publishing Group

Fingerprint Dive into the research topics of 'Inferring introduction routes of invasive species using approximate Bayesian computation on microsatellite data'. Together they form a unique fingerprint.

Cite this