Patterns of geographic variation between mitochondrial and nuclear markers in Heaviside’s (Benguela) dolphins (Cephalorhynchus heavisidii)
JavaScript is disabled for your browser. Some features of this site may not work without it.
Patterns of geographic variation between mitochondrial and nuclear markers in Heaviside’s (Benguela) dolphins (Cephalorhynchus heavisidii)
Gopal, Keshni; Karczmarski, Leszek; Tolley, Krystal A.
Date:
2019-09
Abstract:
The Heaviside's (or Benguela) dolphin (Cephalorhynchus heavisidii) is endemic to the west coast of southern Africa. The present study investigated the population genetic structure across a large portion of the species distribution using mitochondrial control region and nuclear (microsatellite) markers. A total of 395 biopsy skin samples were analyzed; they were collected from free‐ranging Heaviside's dolphins in 7 locations along 1650 km of coast between Table Bay, South Africa and Walvis Bay, Namibia. Both genetic markers rejected the hypothesis of 1 homogenous population but revealed contrasting results in the genetic structuring of putative populations. Mitochondrial DNA suggested either 2 populations or a fine‐scale division with 6 (sub) populations, while microsatellite markers were indicative of 2 widespread populations with measurable gene flow between them. Neutrality tests and mismatch distribution of the mitochondrial sequences indicated a departure from mutation–drift equilibrium due to a population expansion at the 2 extremes of the geographic range, but not towards the middle of the distribution. These results highlight the importance of evaluating multiple genetic markers to gain reliable insights into population processes and structure.
Description:
Supplementary material: Table S1 Summary of genetic variation based on 16 microsatellite loci in the Heaviside's dolphin: Ta stands for annealing temperature, bp for allele sizes, and Na represents the number of alleles examined within each putative population where observed (HO) and expected (HE) heterozygosities were estimated; n indicates the number of individuals used in calculations; dash (−) indicates loci which were not polymorphic, PI stands for Probability of Identity per locus, and PIsibs indicates Probability of Identity for genetic similarity among siblings per locus. Locus SCA22 did not amplify and loci Dde09 and Dde059 were monomorphic
Table S2 Genetic variability estimates in mtDNA control region sequences excluding singletons (n = 27) for haplotype diversity (h) and nucleotide diversity (π)
Table S3 Summary of genetic variation based on 13 microsatellite loci in the Heaviside's dolphin: Na indicates the number of alleles examined per each sampling site where observed (HO) and expected (HE) heterozygosities were estimated, HWE stands for Hardy‐Weinberg equilibrium; n indicates the number of individuals used in calculations, § denotes loci out of HWE (≤0.05) and ∗ indicates evidence for null allele. Locus SCA22 did not amplify and loci Dde09 and Dde059 were monomorphic
Table S4 Sex‐biased dispersal results for males and females with respect to FIS, FST, HO, HS, mean assignment and variance assignment
Table S5 Haplotype frequencies per sampling location of 51 haplotypes identified in biopsy samples of Heaviside's dolphins obtained at seven locations off the southwest African coast
Table S6 STRUCTURE clustering analysis: a. Proportion of Heaviside's dolphin individuals from each of the seven sampling locations assigned to each of the two clusters inferred from STRUCTURE analysis
Figure S1 Median‐joining network of mtDNA control region haplotypes without singletons for Heaviside's dolphins found off the southwest coast of southern Africa.
Figure S2 Average posterior probability (ln K) for each of the seven clusters (i.e., K 1 to 7) from 15 independent runs.
Figure S3 Log‐likelihood values (Ln Pr(X/K)) from 15 independent runs, where the ad hoc statistic delta K (ΔK) shows the most probable number of genetic clusters (K).
Figure S4 Bayesian assignment probabilities where K = 6 clusters were inferred from STRUCTURE analysis. Two sampling sites, Hondeklipbaai and Port Nolloth, are grouped together as in the AMOVA results using mtDNA control region data. Each individual is represented by a thin vertical line, which is portioned into K coloured segments that represent the individual's estimated membership fractions in K clusters. Black lines separate individuals of different populations. The six putative populations are labelled below the figure.
