Computational protocol: Extensive population admixture on drone congregation areas of the giant honeybee, Apis dorsata (Fabricius, 1793)

Similar protocols

Protocol publication

[…] To infer the level of spatial genetic differentiation between the resident colonies and the different DCA locations, we calculated pairwise FST values using the software MSA v. 4.05 (Dieringer and Schlötterer ). To avoid the overrepresentation of siblings in the estimation, only one drone per inferred colony was kept in each DCA location and day of sampling. FST values were similarly calculated to infer temporal genetic differentiation between the 2 days of sampling within each DCA. For both temporal and spatial analyses, significant levels were estimated as the proportion of permuted FST values equal or greater than the observed FST after 10,000 allele permutation for each pair of populations and corrected after Bonferroni. Additionally, Jost's D estimator for genetic differentiation (Jost ) was calculated accordingly for each pair of DCA using the online version of the software SMOGD (Crawford ).To estimate the number of subpopulations contributing to the DCAs and compare them to the drones originating from the resident colonies, the software Structure 2.3.4 (Pritchard et al. ) was utilized based on the sampled drones and derived resident drone genotypes (nonadmixture model; burn-in length: 100,000, run length: 250,000; 25 iterations from K = 1 to K = 8). From the results given by Structure, the delta K value (Evanno et al. ) was inferred using Structure Harvester (Earl and vonHoldt ). The software CLUMPP (Jakobsson and Rosenberg ) was used to find the optimal cluster membership alignment of our Structure analysis over the 25 iterations. These results were plotted using the software Distruct (Rosenberg ).Finally, a Chi-squared test was performed to test the relationship between the inferred subpopulations and the DCA locations and days of sampling. To perform this test, observed proportion of drones per subpopulation in each DCA location and day of sampling were derived from the results of the optimal cluster membership given by the software CLUMPP. The value of the Chi-squared test was calculated manually using a contingency table (subpopulations × DCA location/days). […]

Pipeline specifications

Software tools Structure Harvester, CLUMPP, DISTRUCT
Application Population genetic analysis
Organisms Apis dorsata, Ilex paraguariensis
Diseases Blood Platelet Disorders