Detecting bottlenecks using BOTTLENECK 1.2.02 in wild populations: the importance of the microsatellite structure

Abstract

Reduced, or bottlenecked, populations are more prone to adverse events. Thus, the detection of genetic bottleneck signatures in wildlife is an important issue for conservation. BOTTLENECK 1.2.02 is a software commonly used for detecting genetic characteristics of past bottlenecks. Here we test the efficiency with which this software detects bottlenecks in two koala populations of known history. The sign test performed well for both populations, particularly under the infinite alleles model for mutation. This suggests this model could be the more realistic for marsupial microsatellites than other mutation models. Under the allele frequency distribution test, the two populations falsely appeared to be at mutation/drift equilibrium. However, this test could detect the bottleneck when only imperfect repeat microsatellites were included in the analysis. We thus recommend further investigation of imperfect repeat microsatellites, which could be more powerful for bottleneck detection. These results underline the cautious approach researchers and conservationists should take when studying the past of unknown populations.