Detecting past changes of effective population size
Understanding and predicting population abundance is a major challenge confronting scientists. Several genetic models have been developed using microsatellite markers to estimate the present and ancestral effective population sizes. However, to get an overview on the evolution of population requires...
| Published in: | Evolutionary Applications |
|---|---|
| Main Author: | |
| Other Authors: | |
| Format: | Article in Journal/Newspaper |
| Language: | English |
| Published: |
2014
|
| Subjects: | |
| Online Access: | http://prodinra.inra.fr/ft/B0989C33-3851-4A07-95F0-77777A280127 http://prodinra.inra.fr/record/274834 https://doi.org/10.1111/eva.12170 |
| Summary: | Understanding and predicting population abundance is a major challenge confronting scientists. Several genetic models have been developed using microsatellite markers to estimate the present and ancestral effective population sizes. However, to get an overview on the evolution of population requires that past fluctuation of population size be traceable. To address the question, we developed a new model estimating the past changes of effective population size from microsatellite by resolving coalescence theory and using approximate likelihoods in a Monte Carlo Markov Chain approach. The efficiency of the model and its sensitivity to gene flow and to assumptions on the mutational process were checked using simulated data and analysis. The model was found especially useful to provide evidence of transient changes of population size in the past. The times at which some past demographic events cannot be detected because they are too ancient and the risk that gene flow may suggest the false detection of a bottleneck are discussed considering the distribution of coalescence times. The method was applied on real data sets from several Atlantic salmon populations. The method called VarEff (Variation of Effective size) was implemented in the R package VarEff and is made available at https://qgsp.jouy.inra.fr and at http://cran.r-project.org/web/packages/VarEff. |
|---|