Computational protocol: Population Genetics of Franciscana Dolphins (Pontoporia blainvillei): Introducing a New Population from the Southern Edge of Their Distribution

Similar protocols

Protocol publication

[…] CLUSTALX 2.0.11 [] was used to align DNA sequences and to identify polymorphic sites. The mtDNA haplotypes were compared with those previously published for the species (SA-SK [], L1-L22 [], M1-M19 [], N1 and N3 [], C23-C28 [], CU1-2 and CU4-CU7 []). Haplotypes were verified using DnaSP v5.10.01 []. In order to study patterns of geographical distribution and haplotype relationships, a Median-Joining network [] was implemented in Network (Fluxus Technology Inc.). To remove all superfluous median vectors and links that were not contained in the shortest tree of the network, reducing network complexity, a Maximum-Parsimony post-processing was conducted [].In order to further evaluate the southernmost portion of the species range, we analyzed samples from the localities of NC, CL, MH and RN (). Samples collected from SCL (n = 4), Pinamar (PN) (n = 1) and Bahía Blanca (BB) (n = 3) were not included in the analysis due to the small sample size. Also, genetic data from some studies within FMA IV [,] could not be included as the haplotype frequencies or their exact sample collection site were not reported.Haplotype (h) and nucleotide diversity (π) of the data set were assessed using Arlequin v3.5 [].For the analysis of population structure we performed an Analysis of Molecular Variance (AMOVA) using Arlequin v3.5 []. Since in a previous study [], genetic differences were found between geographically close populations, and considering that the artisanal fisheries in our sampled locations tend not to overlap, we defined 4 populations: NC, CL, MH and RN. Population pairwise FST values were analyzed using Arlequin v3.5 []. We also performed a Mantel test without grouping populations in order to test for isolation by distance (IBD). The correlation was examined between FST/(1-FST) and the logarithm the geographical distance between sites using IBD v3.23 []. Geographical distances between locations, measured as the minimum distance by sea between each other, were calculated using a Geographic Information System (GIS) in ArcGIS software. As evidence of IBD, the rejection of the null hypothesis of a flat or negative slope between genetic and geographical distances was used.In order to study the historical demography of the species, we analyzed the distribution of the observed number of pairwise differences among all haplotypes in a sample, or a mismatch distribution analysis [,] for each of the populations obtained from the population pairwise genetic analysis (see below). Goodness of fit between the observed and expected mismatch was assessed by the Harpending’s raggedness index (r) []. This index quantifies the smoothness of the observed pairwise difference distribution and a nonsignificant result indicates a good fit to a population expansion model []. Typically, populations that had undergone a recent expansion show smooth and unimodal distributions; bimodal distribution patterns are suggestive of two expansions at different times; and populations that had been stationary for a long time show ragged and multimodal distributions [–]. Additionally, Tajima’s D [] and Fu’s F S [] neutrality tests were performed. Both tests were developed to detect departures of DNA polymorphisms from the neutral expectations. Tajima’s D [] uses the frequency of segregating nucleotide sites and Fu’s F S [] uses the haplotypes distribution. Significantly negative values of Tajima’s D [], due to an excess of rare alleles, indicate population expansion or selective sweep, whereas significantly positive values, due to an excess of intermediate frequency alleles, indicate genetic subdivision or diversifying selection. Large negative values of Fu’s F S [], due to an excess of rare alleles, indicate population growth or genetic hitchhiking. All of these analyses were accomplished using DnaSP v5.10.01 [] and Arlequin v3.5 []. Additionally, migration rates and divergence times between putative populations (NC-CL, NC-MH, NC-RN, CL-MH, CL-RN and MH-RN) were obtained with a Markov Chain Monte Carlo (MCMC) approach as implemented in the program MDIV []. The program estimates the parameter theta, which is a product of the effective population size and the mutation rate of the studied gene region (θ = 4N eμ), the migration rate per gene per generation between populations scaled by the effective population size (M = 2N e m), and the time since the two populations diverged scaled by the effective population size (T = t/2N e). We used the finite sites (HKY) model and performed 10 independent runs of 2 x 106 iterations each and a burn-in of 5 x 105 iterations. Likelihood values for each parameter were estimated and those with the highest posterior probability were accepted as the best estimates. We further analyzed patterns of historical demography with the Bayesian skyline plot method of Drummond et al. using BEAST 1.6 []. This model, that uses standard MCMC sampling procedures, provides a powerful framework for estimating effective population size through time. The method produces credibility intervals that represent the combined phylogenetic and coalescent uncertainty []. Coalescent reconstructions used a strict molecular clock with a substitution rate of 1.83 x 10−2 substitutions/site/My [], the HKY+G+I model of mutation, as indicated by JModelTest [], and five grouped intervals. Three replicates of 4 x 107 MCMC steps each were run. The first 10% of each run was discarded as burn-in. Results were checked for convergence to a stationary distribution in Tracer 1.6 and combined using LogCombiner 1.6. Higher estimates of molecular evolutionary rates based on population studies than those inferred from phylogenetic studies have been previously described (e.g. [,]). Since we used a rate estimated from a phylogenetic study for the families of the river dolphins Iniidae and Pontoporiidae [], time estimates for our populations may be overestimated due to the time dependency of molecular evolutionary rates. […]

Pipeline specifications

Software tools Clustal W, DnaSP, Arlequin, BEAST, jModelTest
Applications Phylogenetics, Population genetic analysis
Organisms Pontoporia blainvillei