Computational protocol: Immunogenetic Variation and Differential Pathogen Exposure in Free-Ranging Cheetahs across Namibian Farmlands

Similar protocols

Protocol publication

[…] To examine patterns of sequence variation, nucleotide sequences were edited manually based on their forward and reverse consensus chromatograms using Chromas Pro Version 1.33 (Technelysium Pty Ltd). The sequences were aligned and coding regions were translated into deduced amino acid sequences using Clustal W as implemented in MEGA 3.1 . The MHC-like nature of the sequences was verified through a homology analysis using blastn ( Standard diversity indices were estimated using the software Arlequin 3.1 . Allele frequencies were estimated at all putative loci separately and as a whole haplotype for MHC I loci. For haplotypes, the frequencies of co-occurring alleles (which presumably constitute a haplotype on a given chromosome) were estimated as the number of individual occurrences of a certain allele divided by its total count observed in the population. Expected heterozygosity (HE) was estimated as a general indicator of the amount of genetic variation in the population . Departures from Hardy-Weinberg (H-W) equilibrium were assessed by applying exact tests . The genetic distance between individual alleles were calculated by the number of amino acid substitutions using the Poisson model as implemented in MEGA 3.1 .For the analysis of genetic variation and genetic differentiation at MHC I and II-DRB loci, all samples were classified into north-central or east-central region according to Thalwitzer et al. (). These two regions lack physical barriers but differ substantially in viral pathogen exposure as revealed by seroprevalence studies conducted in Namibian cheetahs , . The ratio of males and females was similar in the north-central (Nmales = 18, Nfemales = 8) and the east-central region (Nmales = 49, Nfemales = 13, χ2 = 0.97, df = 1, p = 0.33). The prediction of different allele and haplotype frequencies between regions was tested by using F-statistics and exact tests of sample differentiation as implemented in Arlequin 3.1. Additionally, we used the chi-square (χ2) and Fisher's method of combined P values obtained by Fisher's exact test as implemented in CHIFISH for MHC I loci. This was done because we combined information from multiple loci which may result in low statistical power and prevent detection of true genetic divergence . This problem occurs particularly in small contingency tables (few populations and few alleles per locus) as observed in our data set. We also tested for differences in the allele and haplotype frequencies between male and female cheetahs to control for any effect due to the male-biased sampling. Differences in heterozygosity between north-central and east-central Namibian cheetahs as well as between males and females were tested with chi-square test in SPSS Version 16.0. […]

Pipeline specifications

Software tools Clustal W, MEGA, BLASTN, Arlequin
Application Population genetic analysis
Organisms Acinonyx jubatus, Viruses