Genetic diversity and demographic history of the largest remaining migratory population of brindled wildebeest (Connochaetes taurinus taurinus) in southern Africa

Abstract

The blue wildebeest (Connochaetes taurinus) is a keystone species in the savannahs of southern Africa, where it maintains shortgrass plains and serves as an important prey source for large carnivores. Despite being the second-largest migratory wildebeest population, the brindled wildebeest (C. t. taurinus) of the Greater Liuwa Ecosystem (GLE) of western Zambia have remained largely unstudied, until recently. While studies have increased understanding of recent demography, migration, and population limiting factors, the level of genetic diversity, patterns of gene flow, and long-term demographic history of brindled wildebeest in the GLE remains unknown. Most genetic studies of wildebeest have focused on small, heavily-managed populations, rather than large, migratory populations of high conservation significance. We used restriction-site associated DNA sequencing (RAD-seq) to assess genetic diversity, population structure, and demographic history of brindled wildebeest in the GLE. Using SNPs from 1,730 loci genotyped across 75 individuals, we found moderate levels of genetic diversity in GLE brindled wildebeest (H_e =  0.210), very low levels of inbreeding (F_IS =  0.033), and an effective population size of about one tenth the estimated population size. No genetic population structure was evident within the GLE. Analyses of the site frequency spectrum found signatures of expansion during the Middle Pleistocene followed by population decline in the Late Pleistocene and early Holocene, a pattern previously observed in other African ungulates. These results will supplement field studies in developing effective conservation plans for wildebeest as they face continued and increasing threats of habitat loss, poaching, and other human impacts across their remaining range.

Introduction

The blue wildebeest (Connochaetes taurinus) is a large antelope that acts as a keystone species in the savannahs of southern and eastern Africa. Many wildebeest populations migrate seasonally in response to precipitation, availability of high-quality forage, and predation risk [1–4]. Grazing by wildebeest maintains shortgrass plains and facilitates nutrient cycling [5–7], which in turn supports communities of other grazing mammals, insects, and birds that depend on small herbaceous plants [8]. Carnivores, such as lions (Panthera leo), hyenas (Crocuta crocuta), wild dogs (Lycaon pictus), and cheetahs (Acinonyx jubatus), often depend on wildebeest as an important prey source [9–12]. Because of the significant ecological roles of wildebeest, changes in migration dynamics or population size could have dramatic impacts on other species and communities in the African grasslands [13].

In many parts of the wildebeest’s range, habitat destruction and fragmentation, obstruction of migration routes, and poaching have impacted wildebeest population size and movement [14]. The conversion of grassland to agricultural fields has reduced available foraging area, contributing to population declines [15,16] and the construction of fences and roads has blocked migration routes and caused previously migratory populations to become sedentary [17,18]. Poaching for bushmeat, driven by deteriorating economic conditions in many communities in and around protected area networks, also threatens wildebeest populations [19–21]. To better understand and address the impact of these threats, there is a need to detect and evaluate their effects on populations.

In addition to experiencing recent declines, wildebeest populations may have undergone fluctuations in the more distant past due to changes in climate or human activity. Genetic analysis of other African ungulates, including African buffalo (Syncerus caffer) [22–24], African elephant (Loxodonta africana) [25], and hippopotamus (Hippopotamus amphibius) [26,27], revealed population declines during the late Pleistocene and early Holocene. These declines coincide with increases in human population size and periods of aridity on the African continent, during which longer dry seasons led to drought and changes in water levels and vegetation types [28–30]. Because many wildebeest populations migrate in response to rainfall and vegetation growth and populations have declined during recent droughts, wildebeest likely experienced similar declines during these two epochs [4,19].

Large population declines during the Late Pleistocene and Holocene may have affected the level of genetic diversity found in the species today, and an understanding of these changes could help in evaluating future demographic effects of climate change. Today, the dryland ecosystems comprising most of the remaining wildebeest range are characterized by extreme seasonality in resource abundance and distribution, making mobility and access to these resources key, particularly in the face of climate change [31]. Understanding how continued habitat fragmentation, loss of migratory routes, and population declines affect genetic diversity and the ability of wildebeest to adapt to the rapidly changing environments of the Anthropocene is therefore of key importance.

The Greater Liuwa Ecosystem (GLE), located in western Zambia and encompassing Liuwa Plain National Park (3,660 km²) and the Upper West Zambezi Game Management Area (22,156 km²; Fig 1), supports a large, migratory population of brindled wildebeest (C. t. taurinus). This flat, open grassland experiences extensive seasonal flooding, which drives the wildebeest to migrate between the drier, southern portion of the GLE, where they spend the wet season (November – April), and the previously flooded northern areas for the dry season (May – October) [32,33]. Though this is thought to be the second largest wildebeest migration in Africa, little was known about this population until recently, after it experienced a significant decline. Poaching, driven by regional politics and civil war in neighboring Angola, led to severe declines of many species in the GLE, including wildebeest, and notably led to the extirpation of lions [32,34]. Additionally, while the wildebeest are thought to have historically migrated between Zambia and Angola, since the Angolan civil war (1975-2002) no transboundary migration has been observed and radio-tracking has shown wildebeest movements to be largely confined to the GLE [33].

[Figure omitted. See PDF.]

(a) Image of a brindled wildebeest (Connochaetes taurinus taurinus; photograph by S. Szarmach). (b) Sampling map depicting where wildebeest tissues were collected (white dots) in the Greater Liuwa Ecosystem (“GLE”). Borders are shown for Liuwa Plain National Park and the surrounding Upper West Zambezi Game Management Area. (c) Location of sampling area (yellow star) within Zambia and broader region. Basemap: NASA Earth Observatory Blue Marble satellite image (obtained through NASA Worldview).

Restoration efforts in Liuwa began in 2003 with the signing of a 20-year management agreement between the government of Zambia and African Parks Network. Ecosystem recovery efforts have resulted in growth and stabilization of the wildebeest population, from an estimated 15,000 individuals in 2001 to 46,000 in 2013 [33–35]. Studies of movement, demography, environmental influences, and predator-prey dynamics [11,32,33,36] have improved understanding of wildebeest population dynamics in the GLE, but the amount of genetic diversity present in the population and its long-term demographic history remain unknown. Past changes in population size have important implications for conservation in the present day, because historical bottlenecks can reduce genetic diversity and limit future adaptive potential, and the comparison of present day to past effective population sizes can improve understanding of current conservation risks for the population [37,38].

Despite the wildebeest being an iconic species of the African savannahs with high conservation significance, until very recently [39], few genetic studies of wildebeest had been published. Studies utilizing molecular markers can inform wildlife management by delineating conservation units, revealing inbreeding, and divulging historical patterns of movement and demographic change [40,41]. Most previous genetic studies of wildebeest focused on heavily managed populations in South Africa, and few evaluated the potential impacts of habitat loss, fragmentation, and population declines occurring across remaining wildebeest range. Early studies assessing wildebeest genetic diversity used allozymes, microsatellites or mitochondrial DNA primarily to inform management of small populations of wildebeest on game ranches in South Africa [42–44]. Microsatellite and mtDNA studies that included samples from a greater part of the blue wildebeest’s range identified high levels of genetic divergence between wildebeest in southern versus eastern Africa, as well as lower levels of significant differentiation between pairs of populations within these regions [45,46].

More recently, whole-genome resequencing of blue wildebeest from much of the species’ range found significant genetic differentiation between all subspecies of blue wildebeest, as well as between local populations of the western white-bearded (C. t. mearnsi) subspecies in eastern Africa and the brindled subspecies in southern Africa [39]. Importantly, Liu et al. [39] found that wildebeest populations where migration had recently been disrupted showed lower genetic diversity and higher levels of inbreeding compared to fully migratory populations. No previous genetic studies have included wildebeest from the GLE, despite this being one of the largest migratory wildebeest populations. The effects of past changes in population size, both historical and recent, on genetic diversity in this recovering migratory population remain unknown.

We assessed genetic diversity, population structure, and demographic history of the GLE brindled wildebeest population using a genome-wide SNP dataset generated through restriction-site associated DNA sequencing (RAD-seq). We first evaluated the current level of genetic diversity present in the GLE wildebeest population, estimated contemporary effective population size, and assessed whether genetic population structure is present. We used both model-based (STRUCTURE) and model-free (PCA) approaches to determine whether multiple populations exhibiting significant differences in SNP allele frequencies are present within the GLE, or whether this region is occupied by a single, panmictic population. We then inferred the past demographic history of the GLE brindled wildebeest population and tested the hypothesis that genetic signatures of population contraction are present due to past climate change and anthropogenic pressures. To do this, we tested demographic models using the coalescent simulator fastsimcoal2 [47] and implemented the model-independent stairway plot method [48,49] to estimate effective population size over time based on the site frequency spectrum.

Materials and methods

Sample collection and DNA extraction

As part of long-term field studies conducted with the Zambia Department of National Parks and Wildlife (DNPW) and African Parks (e.g., [33]), we collected 97 tissue samples in the GLE between 2011−2014 from wildebeest that were either anesthetized for radio collaring or from opportunistically sampled wildebeest carcasses detected during the course of fieldwork. Wildebeest immobilized for radio collaring were remotely injected with M99 (etorphine) administered using cartridge-fired dart rifles projected from vehicle or helicopter. Once the animal was fully immobilized and unable to feel pain, a tissue sample was collected from either a small ear notch or the removal of a small amount of skin tissue from the hindquarters. A small incision was made on the skin using a sterile scalpel, tissue extracted and immediately stored in ethanol, and the sampling site was disinfected and treated. Temperature, respiration, and general condition were monitored throughout the period when the animal was anaesthetized, and the animal was blindfolded during immobilization to reduce stimuli. Each animal was immobilized long enough for a collar to be fit, teeth to be measured, and a tissue sample taken before a complete reversal with M5050 (diprenorphine) and naltrexone, typically 15-20 minutes total duration from induction to recovery. After reversal, the animal was monitored for recovery. As all immobilized animals were part of a long-term study, we were able to monitor each wildebeest long-term and observed no adverse effects from immobilization or sample collection. All animal handling procedures were performed following animal welfare standards and protocols required by the Zambia Department of Veterinary and Livestock Services and the DNPW, and a Zambian-registered veterinarian conducted and oversaw all immobilization operations. Tissue samples obtained from carcasses were largely from carnivore predations that were detected opportunistically or in the course of carnivore hunt follows [12] and followed similar collection and storage protocols.

Samples were exported with Zambia Wildlife Authority Permit #7830 and imported with USDA Veterinary Permit #109174. We extracted genomic DNA from the tissues using the DNeasy Blood & Tissue Kit (Qiagen, Valencia, CA, USA) and quantified the concentration and purity of the extracted DNA using a Nanodrop 2000c. DNA extractions with a concentration < 10 ng/μL were combined with DNA from a second extraction from the same tissue sample, if available, and concentrated using vacuum centrifugation.

RAD library preparation and sequencing

We prepared double-digest restriction site associated DNA (ddRAD) libraries following the protocol of DaCosta and Sorenson [50]. First, ~ 1 µg of genomic DNA from each wildebeest was digested using the restriction enzymes EcoRI-HF and SbfI-HF (New England Biolabs, Ipswich, MA, USA). Unique combinations of barcoded adapters were ligated to the digested DNA from each individual, so that samples could later be pooled for sequencing. Restriction fragments within a size range of 200 − 400 bp were selected through gel excision, and DNA was purified using a Qiagen MinElute Gel Extraction Kit. The size-selected DNA fragments were PCR amplified using Phusion High-Fidelity PCR Master Mix (Thermo Scientific, Waltham, MA, USA). PCR products were purified with AMPure XP SPRI beads (Beckman Coulter, Brea, CA, USA). The concentrations of the purified PCR products were quantified using real-time PCR, and samples were pooled in equimolar amounts, with the exception of some low-concentration samples, which were not included in the final pool. Ultimately, 83 samples were included in the multiplexed library, which was sequenced on one lane of an Illumina HiSeq 2500 Sequencing System (Tufts University, Medford, MA, USA), generating single-end reads of 100 bp in length.

Alignment, genotyping, and quality filtering

After sequencing, we used FastQC version 0.11.7 to assess per-base sequence quality for the reads [51]. We used a computational pipeline developed by DaCosta and Sorenson [50] for analyzing ddRAD-seq data to process the raw sequence reads (Python scripts available at https://github.com/BU-RAD-seq/ddRAD-seq-Pipeline). We used the script ddRADparser.py to correct errors in restriction sites and barcodes and to demultiplex the reads, assigning each read to an individual wildebeest based on the combination of barcodes associated with the read. The demultiplexed sequence data are available from the NCBI Sequence Read Archive (SRA) under the following accession numbers: BioProject ID PRJNA588629; BioSample IDs SAMN13254976 − SAMN13255074. After demultiplexing, identical sequences were collapsed and reads with low average Phred quality scores (Q <  20) were filtered out. The condensed and filtered reads were then clustered into putative loci across individuals based on a similarity threshold of 0.85 using the UCLUST algorithm in the program USEARCH version 5 [52]. A representative sequence from each cluster was aligned to a reference genome using the blastn program in BLAST version 2.8.0 [53] and the following settings: evalue =  0.0001, word_size =  11, gapopen =  5, gapextend =  2, penalty =  -3, reward =  1, and dust =  yes). Sequences were aligned to two reference genomes: v4 of the chromosome-level domestic sheep (Ovis aries) genome ([54]; NCBI Accession #GCA_000298735.2) and the scaffold-level blue wildebeest genome ([55]; NCBI Accession #GCA_006408615.1). Separate clusters that generated BLAST hits to the same genomic location were combined into a single cluster, and those that did not generate hits to the reference genome were retained but marked as anonymous loci. Reads were aligned within each cluster using MUSCLE version 3.8.31 [56].

We genotyped individuals at the identified loci using DaCosta and Sorenson’s RADGenotypes.py script [50], which calls individuals as homozygous for a SNP if > 93% of reads exhibit the same allele, and as heterozygous if > 29% of reads show a second allele. When a lower proportion of reads exhibited a minor allele, the genotype was flagged as a “provisional heterozygote” (20-29% of reads) or “ambiguous” (7-20%) and the locus was manually checked in MEGA [57] for signs of misalignment, such as large gaps in the alignment or long series of successive SNPs, before being included in the final set of SNPs. These thresholds are similar to standardly accepted cutoffs for genotyping SNPs in high-coverage sequence data [58], and have been used to successfully genotype SNPs in other RAD-seq datasets [59–62]. A custom R script utilizing the tidyverse collection of packages [63,64] was used to remove loci with substantial amounts of missing data and putative “paralogous loci” formed by erroneous clustering of sequences originating from repetitive elements in the genome (R script available at https://github.com/sszarmach/NMU-wildebeest). Samples with an average sequencing depth of < 10 reads per locus and loci with < 5 reads in > 10% of samples were removed, since low sequencing depth can lead to genotyping errors [65]. We filtered out loci that significantly departed from Hardy-Weinberg equilibrium based on a chi-square goodness-of-fit test (α =  0.05), because such deviation can result from null alleles (heterozygote deficiency) or incorrectly clustered paralogous loci (heterozygote excess). To eliminate putative paralogous loci, we also removed clusters that exhibited more than two alleles, that had abnormally high read depths (mean depth > 700), that had > 4 SNPs, or that generated multiple high-quality BLAST hits to the reference genome [66,67]. For all downstream analyses except inference of demographic history using the site frequency spectrum, a minor allele frequency filter (MAF > 0.01) was applied to remove loci containing singleton SNPs, which could be an artifact of sequencing error. These quality filtering steps are outlined in S1 Table.

A final data set of 1,730 polymorphic loci genotyped across 75 individuals was used in downstream population genetic analyses. We randomly drew one SNP from each RAD locus to avoid the inclusion of multiple SNPs that were linked due to chromosomal proximity. We used PGDSpider version 2.1.1.5 to convert the SNP data file into the correct format for each analysis program [68].

Population structure

We first explored whether genetic structure is present within the GLE wildebeest population using the model-based clustering approach implemented in STRUCTURE version 2.3.4 [69–72]. We assumed the admixture model and correlated allele frequencies [70,73]. Ten independent runs were performed for each number of genetic clusters tested (K =  1 to K =  4), and each run contained 10,000 burn-in steps and 100,000 MCMC steps. We used STRUCTURE HARVESTER web version 0.6.94 [74,75] to summarize STRUCTURE results, and the best K value was selected based on the log probability of the data given the number of clusters (ln Pr(X | K)) and the Evanno ΔK method [76]. Visualizations of the STRUCTURE output were created using the R package and Shiny app pophelper [77]. We also examined genetic structure using a non-model-based approach by performing a principal components analysis (PCA) in the R package adegenet [78].

Genetic diversity

After determining no genetic population structure was present within the GLE, we calculated allele frequencies and expected and observed heterozygosities across all samples using the R package adegenet [78,79]. We tested SNPs for deviation from Hardy-Weinberg equilibrium using the hw.test function in the R package pegas (Fisher’s exact test with 1,000 MCMC replicates) [80]. We corrected P-values from the Hardy-Weinberg tests for multiple comparisons using the Benjamini and Hochberg method in the R package sgof [81–83]. The inbreeding coefficient, F_IS, was estimated using the R package hierfstat [84].

Estimation of effective population size

We estimated contemporary effective population size (N_e) for GLE brindled wildebeest using the bias-corrected linkage disequilibrium method [85,86] implemented in NeEstimator version 2.1 [87] assuming a random-mating model. We constructed 95% confidence intervals for N_e using the parametric method of Waples [85] and the jackknifing method of Jones et al. [88], which accounts for pseudoreplication resulting from physically linked loci and is recommended for data sets including large numbers of SNPs.

Demographic history

Using the coalescent simulator fastsimcoal version 2.6 [47] we evaluated five models of historical demographic change: constant population size, expansion, decline, bottleneck, and expansion followed by decline (hereafter “expansion-decline”; Fig 2). Fastsimcoal2 simulates the expected site frequency spectrum (SFS) for a demographic model provided by the user and computes the composite likelihood of the SFS observed from the data given this model. We computed the folded site frequency spectrum from the minor allele frequencies of all sites in 1,831 loci (2,946 SNPs and 172,603 monomorphic sites) that passed quality filtering without applying the minor allele frequency filter, as excluding singletons can bias inferences drawn from the site frequency spectrum [89].

[Figure omitted. See PDF.]

For each model, parameters describing effective population size (N) and select times before present (T) were estimated.

For each model, we ran 50 replicate simulations, each with randomly selected initial parameter values, to increase the likelihood that the global maximum solution was discovered. Each run included 100,000 coalescent simulations and 40 cycles of Brent’s ECM algorithm, which is used to select parameters for the model that optimize the composite likelihood [47]. We provided broad parameter search ranges of 10–100,000 gene copies (haploid N) for all N_e estimates, except N_ancestral in the decline and bottleneck models and N_mid in the expansion-decline model, which were set to 10–1,000,000 gene copies. The search range for all event time parameters was 10-100,000 generations. The mutation rate was set to 1.7x10^-8 substitutions per base pair per generation, based on the average mammalian mutation rate (2.22x10^-9 substitutions per base pair per year) [90] and the generation time of wildebeest (7.5 years) [91]. For each demographic scenario, the replicate with the highest likelihood was retained as the best model. Fastsimcoal2 handles effective population size as the number of gene copies, so N_e was converted to the number of diploid individuals by dividing all estimated N_e parameters by two. Time parameters were converted from number of generations to years before present by multiplying T estimates by a generation time of 7.5 years. We calculated Akaike Information Criterion (AIC) values [92] for the models with the highest maximum composite likelihood for each demographic scenario and compared the demographic models using Akaike weights (AICw) calculated using the akaike.weights function in the qpcR R package [93].

In addition to the model-based approach implemented in fastsimcoal2, we also inferred historical changes in population size using the model-independent stairway plot method. Stairway Plot 2 uses a flexible multi-epoch model to estimate population size over different time intervals based on the site frequency spectrum [48,49]. The same folded site frequency spectrum, mutation rate, and generation time used in the fastsimcoal2 analysis were used to generate the stairway plot. Confidence intervals were created based on 200 bootstrapped site frequency spectra drawn from the observed SFS.

Results

Sequencing produced ~ 220 million reads that passed initial Illumina quality filtering, and demultiplexing resulted in an average of 2.2 million reads assigned to each wildebeest. After these reads were filtered for quality and clustered by similarity, 22,483 RAD loci were identified. A similar number of loci aligned to each reference genome: 19,216 (85.5%) of sequences generated at least one BLAST hit to the sheep genome, and 19,267 (85.7%) of loci aligned to the blue wildebeest genome. Eight wildebeest samples were removed due to low sequencing depth (<10 reads per locus), leaving 75 wildebeest that were genotyped. There were 3,417 polymorphic loci that were genotyped in > 90% of samples. Many putatively paralogous loci were removed from the dataset, likely the result of the large proportion of repetitive sequence in bovid genomes (~50%) [55,94]. After all quality filtering (S1 Table), the final dataset included 1,730 high-quality polymorphic RAD loci (mean length =  96 bp, mean depth per locus per individual =  355X) that passed a minor allele frequency (MAF) threshold of 0.01, containing 2,921 SNPs. The number of loci that aligned to each chromosome of the sheep genome was correlated with the length of the chromosome (Spearman’s ρ =  0.683, P <  0.001; S1 Fig), indicating that the dataset contains loci evenly spaced throughout the genome. We estimated population structure, genetic diversity and effective population size for GLE brindled wildebeest (n =  75) using 1,730 biallelic SNPs, with one SNP randomly drawn from each RAD locus.

We did not detect any genetic population structure within the brindled wildebeest of the Greater Liuwa Ecosystem. The most likely number of clusters identified using STRUCTURE was K =  1 (S2 Table), and for each K value > 1 that was tested, almost all individuals were assigned membership to multiple clusters (S2 Fig). Principal component analysis also did not show any genetic clustering of subpopulations within the GLE (Fig 3). Rather, the GLE brindled wildebeest seem to compose a single, panmictic population.

[Figure omitted. See PDF.]

Principal component analysis (PCA) of GLE brindled wildebeest (n =  75) based on 1,730 SNPs.

After correcting for multiple comparisons, no SNPs in the final dataset significantly deviated from Hardy-Weinberg equilibrium (α =  0.05). The minor allele frequency per SNP ranged from 0.01 to 0.50, with a mean of 0.14 (SD =  0.13). The observed heterozygosity of the SNPs ranged from 0 to 0.597, and the average observed heterozygosity of GLE brindled wildebeest was 0.204 (SD =  0.150; Fig 4). Expected heterozygosity per SNP ranged from 0.013 to 0.504, and the average expected heterozygosity was 0.209 (SD =  0.151). The inbreeding coefficient (F_IS) for the population was 0.031 (95% CI: 0.026 – 0.038). The estimated effective population size (N_e) was 5,075 (parametric 95% CI =  3,360–10,322; jackknifing 95% CI =  800–∞).

[Figure omitted. See PDF.]

Boxplots show median and interquartile range.

The results of both the model-based fastsimcoal2 analysis and model-independent stairway plot analysis suggest that the GLE brindled wildebeest experienced population expansion during the Pleistocene followed by large declines throughout the Late Pleistocene and Holocene. Of the five demographic models simulated in fastsimcoal2 (Fig 2), the expansion-decline model had the highest composite likelihood score and Akaike weight (w_i; Table 1) and the expected site frequency spectrum for this model was the most similar to the observed SFS (S3 Fig). The parameters of the highest-likelihood expansion-decline model suggest a population increase from N_e = 26,800 to N_e = 92,700 occurring 635,000 years ago, followed by a population decline to N_e = 13,400 occurring 15,480 years ago (S3 Table). The stairway plot exhibited similar changes in population size as the fastsimcoal2 expansion-decline model (Fig 5), with an increase in population size from N_e = 30,000 to N_e = 90,000 from 1,000,000-500,000 years ago, followed by a sharp decline to N_e = 11,000 from 25,000-3,000 years ago.

[Figure omitted. See PDF.]

[Figure omitted. See PDF.]

The red line shows the median estimated effective population size, the dark grey lines encompass the 75% CI, and the light gray lines contain the 95% CI, determined from 200 bootstrapped site frequency spectrums.

Discussion

Our results show that a single, panmictic brindled wildebeest population occupies the GLE, possessing moderate levels of genetic diversity, very low levels of inbreeding, and no genetic population structure. Though few population genomic studies of wild African ungulates have measured SNP-based heterozygosity, the level of heterozygosity seen in GLE brindled wildebeest (average per-SNP H_o =  0.204) was slightly higher than average per-SNP heterozygosity in another African antelope, the springbok (Antidorcas marsupialis; H_o =  0.15) [95]. One other study has reported per-SNP heterozygosity for reduced-representation genomic data from brindled wildebeest (~20,000 SNPs generated using DArT-seq) [96]. From their dataset generated for brindled wildebeest from a South African private game ranch, van Deventer et al. [96] found a mean minor allele frequency (0.19) and average observed heterozygosity (0.19) similar to those of the GLE brindled wildebeest. However, in the captive South African population of van Deventer et al. [96], observed heterozygosity was significantly lower than expected heterozygosity (0.27), potentially a sign of inbreeding that was not seen in the wild GLE brindled wildebeest studied here.

While direct comparison to other previously published studies of genetic variation in wildebeest is more challenging, especially those that utilized allozymes, microsatellites, and mitochondrial DNA, the GLE brindled wildebeest seem to have relatively greater genetic diversity than some previously studied populations. Researchers found low allozyme polymorphism and low heterozygosity in wildebeest from Tanzania and Kenya [97], and low mtDNA nucleotide diversity [42] and allozyme heterozygosity [43] in South African wildebeest. Many of these early studies focused on smaller, more heavily managed wildebeest populations that likely were experiencing founder effects resulting from translocations of individuals from one reserve to another. In contrast, brindled wildebeest in the GLE are free-ranging and do not have a history of translocations or intensive management. In a recent study of black (C. gnou) and blue wildebeest populations throughout their range using whole-genome data, Liu et al. [39] found lower genome-wide heterozygosity estimates and more runs of homozygosity (a sign of past inbreeding) in nonmigratory wildebeest populations compared to migratory ones. Though the GLE wildebeest numbers declined drastically in the early 2000’s and the migration no longer extends into Angola, it does not seem that these changes have resulted in low genetic diversity in Liuwa compared to other populations.

The GLE brindled wildebeest exhibited a very low signature of inbreeding. Though F_IS was significantly greater than zero (0.033; 95% CI: 0.026 – 0.039), indicating a slight excess of homozygosity, this low value does not indicate extensive nonrandom mating in the population. Similar values of the inbreeding coefficient have been seen in springbok (F_IS =  0.018 – 0.032) from several populations in South Africa [95]. While the GLE brindled wildebeest population experienced significant declines in the early 2000’s [34], it seems these declines were not great enough and did not last long enough (occurring over a few generations) to lead to inbreeding and reduction of genetic diversity in the population. We also found no evidence of population structure using either STRUCTURE or PCA, indicating that a single, panmictic wildebeest population inhabits the GLE. Visual observation and GPS tracking of collared wildebeest have found that the GLE wildebeest spend the wet season in a concentrated area in the southern portion of the park [33], and our genetic results corroborate field observations suggesting that there is likely no spatial segregation of individuals during breeding.

Our estimate of the current effective population size (N_e), the size of an idealized population that would experience the same amount of genetic drift as the population being studied [98], was 5,075 (95% CI =  3,360–10,322). Estimating a population’s effective size is important for conservation, because a population with low N_e could lose genetic diversity faster than expected based on its census population size (N). The effective population size estimated for wildebeest in the GLE is 11% of the total estimated population size determined using aerial surveys in 2013 (N =  46,000) [35]—an effective to census population size ratio (N_e/N) similar to estimates for many comparable mammals. Reviews have found the average N_e/N ratio in wildlife to be 0.10-0.14 [99,100]. In African buffalo, another large, highly mobile ungulate with a promiscuous mating system, researchers have estimated N_e/N ratios ranging from 0.10 to 0.20 [101]. When also considering the moderate level of heterozygosity and low inbreeding coefficient for the Liuwa population, it seems unlikely that this estimate of current effective population size is indicative of a loss in genetic diversity after the population decline in the early 2000’s. Very recent population declines can actually increase the N_e/N ratio, because census population size decreases faster than the timescale required for N_e to be reduced by a loss of genetic diversity [37,38,100].

Using both model-based and model-independent methods, we detected signatures of population expansion followed by decline in the site frequency spectrum of GLE brindled wildebeest. Both the highest-likelihood expansion-decline fastsimcoal2 model and the stairway plot inferred a threefold increase in population size during the Middle Pleistocene, followed by an ~ 87% reduction in population size during the Late Pleistocene and early Holocene. Evidence of Pleistocene population expansions followed by Holocene declines has been found in a number of other African mammal species using genetic data, including African buffalo [23,24], hippopotamus [26, 27], African elephant [25,102], and savannah baboon [103]. In a comparative study of ruminant whole genomes, including 26 African species, Chen et al. [55] found using the pairwise sequentially Markovian coalescent (PSMC) method that most of the African species shared a similar demographic history of population expansion followed by decline in the late Pleistocene. Similarly, Liu et al. [39] used PSMC to infer effective population size over time for the five blue wildebeest subspecies, and the demographic history inferred for the brindled wildebeest (which did not include any individuals from the GLE) is very similar to the results of our stairway plot analysis in both the timing of the expansion and decline and the effective populations sizes inferred.

Two primary explanations have been proposed to explain the population declines in African ungulates during the late Pleistocene and early Holocene: climatic change and anthropogenic impacts [104]. Ice core records and other paleontological data suggest that Africa experienced large fluctuations in precipitation during the late Pleistocene and early Holocene, with extensive periods of drought [28–30]. Such climatic changes could have a profound impact on wildebeest, as modern-day wildebeest populations have fluctuated in response to drought [19] and wildebeest migration is at least partially driven by changes in rainfall and forage availability [4]. However, others have argued that climatic changes in the late Pleistocene were no more extreme than those occurring earlier in the epoch, and thus do not alone explain population declines and extinctions [104]. Human population growth and the advent of pastoralism during the Neolithic revolution may have increased pressures on ungulate populations already stressed by drought, leading to the observed declines [23,55]. Past population declines in the face of climatic changes and intensified human activity reflect the modern-day pressures faced by wildebeest, as habitat destruction, overhunting, and climate change all threaten the species today.

Our results show that—despite recent declines in population size and the disruption of migration to Angola—the GLE brindled wildebeest population does not seem to have experienced recent reductions in genetic diversity or increased inbreeding. The ecosystem restoration efforts implemented in the GLE likely allowed the wildebeest population to increase before the bottleneck became severe enough to result in detectable reductions in genetic diversity. While the analyzed samples were collected ~ 10 years ago, recent surveys have not detected large changes in population size since, so we do not expect patterns of genetic diversity to have changed significantly during the intervening time [33]. However, wildebeest continue to face threats of habitat destruction, obstruction of migration routes from fencing, and climate change, and continued genetic monitoring could detect whether these impacts are influencing genetic diversity in the GLE in the future. This is especially important given the evidence that brindled wildebeest population size drastically decreased during past periods of climate change and anthropogenic pressure in the late Pleistocene and early Holocene.

Future sampling and genomic analysis of wildebeest from Kafue National Park in Zambia could help to elucidate patterns of connectivity or differentiation between brindled wildebeest in the GLE and those in a neighboring population that is also of conservation significance. The Zambezi River, which separates these two protected areas and currently acts as a barrier, may have historically prevented gene flow, leading to genetic differentiation between these populations. Analysis of four brindled wildebeest genomes from Kafue by Liu et al. [39] suggested that the Kafue population was genetically distinct from more southern brindled wildebeest populations, but the relationship between the Kafue and GLE populations remains unknown. Understanding patterns of genetic differentiation between GLE brindled wildebeest and those in Kafue National Park would have important implications for future management decisions such as translocations between these populations. Our study provides a baseline understanding of genetic diversity and long-term demographic history for Africa’s second largest migratory wildebeest population, which will assist the development of effective conservation plans for the species as wildebeest face threats of habitat destruction and overhunting.

Supporting information

S1 Fig. Correlation between the number of RAD loci that aligned to each chromosome of the domestic sheep (Ovis aries) genome and the length of the chromosome.

The linear regression line is shown in grey. The number of RAD loci mapping to a chromosome was significantly correlated with chromosome length (Spearman’s ρ =  0.683, P =  0.00012).

https://doi.org/10.1371/journal.pone.0310580.s001

S2 Fig. Structure plots for 75 brindled wildebeest from the Greater Liuwa Ecosystem based on 1,730 SNPs assuming K = 2, K = 3, and K = 4 clusters.

None of these K values had a higher likelihood than K = 1, and the assignment of most individuals to multiple clusters indicates that there is no genetic population structure within the GLE.

https://doi.org/10.1371/journal.pone.0310580.s002

S3 Fig. Comparisons of the folded site frequency spectrum observed from 1,730 SNPs from Greater Liuwa Ecosystem brindled wildebeest (black) and expected site frequency spectra simulated in fastsimcoal2 for five demographic models (colored).

(a) Constant population size, (b) population expansion, (c) population decline, (d) bottleneck, (e) population expansion followed by decline.

https://doi.org/10.1371/journal.pone.0310580.s003

S1 Table. Number of putative RAD loci remaining after each step of the quality filtering protocol.

Data is presented separately for loci aligned to the domestic sheep (Ovis aries) or brindled wildebeest (Connochaetes taurinus) reference genomes. Loci aligned to the wildebeest genome were used for downstream analyses.

https://doi.org/10.1371/journal.pone.0310580.s004

S2 Table. Statistics obtained from Structure Harvester used to determine the optimal number of genetic clusters for wildebeest within the Greater Liuwa Ecosystem.

For each K value (number of genetic clusters assumed in the model), we report the number of replicates, mean and standard deviation log probability of the data given the number of clusters, and Evanno statistic, ΔK.

https://doi.org/10.1371/journal.pone.0310580.s005

S3 Table. Parameter estimates from the highest likelihood model for each demographic scenario tested in fastsimcoal2.

Parameters describe effective population size (N) or the time a demographic event occurred (T) in years before present. Estimates have been rounded to the nearest hundred. The model with the highest support is shown in bold.

https://doi.org/10.1371/journal.pone.0310580.s006

S1 File. Sampling Information Supplementary File.

Table containing information for all wildebeest with RAD-seq data, including sample collection date, latitude, longitude, age, and sex.

https://doi.org/10.1371/journal.pone.0310580.s007

(XLSX)

Acknowledgments

Our thanks to the Zambia Department of National Parks and Wildlife, African Parks Network, and the Barotse Royal Establishment for permission and collaboration in this research. We gratefully acknowledge the late Jeffrey Horn of the NMU Computer Science Department, who passed away in February 2024. Jeff provided the computing infrastructure and assistance that made the data analysis for this work possible. We appreciate and miss his enthusiasm for collaborating across disciplines. We also thank Grant Combs for computing assistance. We are very grateful to Michael Sorenson and Jeffrey DaCosta for guidance on the RAD-seq protocol and analysis pipeline. We also thank Kurt Galbreath for input and advice on the project and Catherine Sun for feedback on the manuscript.

References

1. 1. Fryxell JM, Sinclair AR. Causes and consequences of migration by large herbivores. Trends Ecol Evol. 1988;3(9):237–41. pmid:21227239

* View Article

* PubMed/NCBI

* Google Scholar

2. 2. Wilmshurst J, Fryxell J, Farm B, Sinclair A, Henschel C. Spatial distribution of Serengeti wildebeest in relation to resources. Journal Title. 1999;7710.

* View Article

* Google Scholar

3. 3. Boone RB, Thirgood SJ, Hopcraft JGC. Serengeti wildebeest migratory patterns modeled from rainfall and new vegetation growth. Ecology. 2006;87(8):1987–94. pmid:16937638

* View Article

* PubMed/NCBI

* Google Scholar

4. 4. Holdo RM, Holt RD, Fryxell JM. Opposing rainfall and plant nutritional gradients best explain the wildebeest migration in the Serengeti. Am Nat. 2009;173(4):431–45. pmid:19243258

* View Article

* PubMed/NCBI

* Google Scholar

5. 5. McNaughton SJ. Serengeti migratory wildebeest: facilitation of energy flow by grazing. Science. 1976;191(4222):92–4. pmid:17834943

* View Article

* PubMed/NCBI

* Google Scholar

6. 6. Holdo RM, Holt RD, Coughenour MB, Ritchie ME. Plant productivity and soil nitrogen as a function of grazing, migration and fire in an African savanna. Journal of Ecology. 2006;95(1):115–28.

* View Article

* Google Scholar

7. 7. Estes RD, Raghunathan TE, van Vleck D. The Impact of Horning by Wildebeest on Woody Vegetation of the Serengeti Ecosystem. J Wildl Manage. 2008;72(7):1572.

* View Article

* Google Scholar

8. 8. Sinclair ARE. Mammal population regulation, keystone processes and ecosystem dynamics. Philos Trans R Soc Lond B Biol Sci. 2003;358(1438):1729–40. pmid:14561329

* View Article

* PubMed/NCBI

* Google Scholar

9. 9. Hayward MW, Kerley GIH. Prey preferences of the lion (Panthera leo). Journal of Zoology. 2005;267(3):309–22.

* View Article

* Google Scholar

10. 10. Hayward MW. Prey preferences of the spotted hyaena (Crocuta crocuta) and degree of dietary overlap with the lion (Panthera leo). Journal of Zoology. 2006;270(4):606–14.

* View Article

* Google Scholar

11. 11. Creel S, Dröge E, M’soka J, Smit D, Becker M, Christianson D, et al. The relationship between direct predation and antipredator responses: a test with multiple predators and multiple prey. Ecology. 2017;98(8):2081–92. pmid:28475209

* View Article

* PubMed/NCBI

* Google Scholar

12. 12. Dröge E, Creel S, Becker MS, M’soka J. Spatial and temporal avoidance of risk within a large carnivore guild. Ecol Evol. 2016;7(1):189–99. pmid:28070283

* View Article

* PubMed/NCBI

* Google Scholar

13. 13. Lee DE, Kissui BM, Kiwango YA, Bond ML. Migratory herds of wildebeests and zebras indirectly affect calf survival of giraffes. Ecol Evol. 2016;6(23):8402–11. pmid:28031792

* View Article

* PubMed/NCBI

* Google Scholar

14. 14. Estes R. The gnu’s world: Serengeti wildebeest ecology and life history. Berkeley: University of California Press; 2014.

15. 15. Serneels S, Lambin EF. Impact of land‐use changes on the wildebeest migration in the northern part of the Serengeti–Mara ecosystem. Journal of Biogeography. 2001;28(3):391–407.

* View Article

* Google Scholar

16. 16. Homewood K, Lambin EF, Coast E, Kariuki A, Kikula I, Kivelia J, et al. Long-term changes in Serengeti-Mara wildebeest and land cover: pastoralism, population, or policies?. Proc Natl Acad Sci U S A. 2001;98(22):12544–9. pmid:11675492

* View Article

* PubMed/NCBI

* Google Scholar

17. 17. Whyte C, Joubert I. Blue wildebeest population trends in the Kruger National Park and the effects of fencing. South African Journal of Wildlife Research. 1988;18(1):78–87.

* View Article

* Google Scholar

18. 18. Holdo RM, Fryxell JM, Sinclair ARE, Dobson A, Holt RD. Predicted impact of barriers to migration on the Serengeti wildebeest population. PLoS One. 2011;6(1):e16370. pmid:21283536

* View Article

* PubMed/NCBI

* Google Scholar

19. 19. Mduma SAR, Sinclair ARE, Hilborn R. Food regulates the Serengeti wildebeest: a 40‐year record. Journal of Animal Ecology. 1999;68(6):1101–22.

* View Article

* Google Scholar

20. 20. Thirgood S, Mosser A, Tham S, Hopcraft G, Mwangomo E, Mlengeya T, et al. Can parks protect migratory ungulates? The case of the Serengeti wildebeest. Animal Conservation. 2004;7(2):113–20.

* View Article

* Google Scholar

21. 21. Harris G, Thirgood S, Hopcraft J, Cromsight J, Berger J. Global decline in aggregated migrations of large terrestrial mammals. Endang Species Res. 2009;755–76.

* View Article

* Google Scholar

22. 22. Heller R, Lorenzen ED, Okello JBA, Masembe C, Siegismund HR. Mid-Holocene decline in African buffalos inferred from Bayesian coalescent-based analyses of microsatellites and mitochondrial DNA. Mol Ecol. 2008;17(22):4845–58. pmid:19140976

* View Article

* PubMed/NCBI

* Google Scholar

23. 23. Heller R, Brüniche-Olsen A, Siegismund HR. Cape buffalo mitogenomics reveals a Holocene shift in the African human-megafauna dynamics. Mol Ecol. 2012;21(16):3947–59. pmid:22725969

* View Article

* PubMed/NCBI

* Google Scholar

24. 24. de Jager D, Glanzmann B, Möller M, Hoal E, van Helden P, Harper C, et al. High diversity, inbreeding and a dynamic Pleistocene demographic history revealed by African buffalo genomes. Sci Rep. 2021;11(1):4540. pmid:33633171

* View Article

* PubMed/NCBI

* Google Scholar

25. 25. Okello JBA, Wittemyer G, Rasmussen HB, Arctander P, Nyakaana S, Douglas-Hamilton I, et al. Effective population size dynamics reveal impacts of historic climatic events and recent anthropogenic pressure in African elephants. Mol Ecol. 2008;17(17):3788–99. pmid:18643879

* View Article

* PubMed/NCBI

* Google Scholar

26. 26. Okello JBA, Nyakaana S, Masembe C, Siegismund HR, Arctander P. Mitochondrial DNA variation of the common hippopotamus: evidence for a recent population expansion. Heredity (Edinb). 2005;95(3):206–15. pmid:16030528

* View Article

* PubMed/NCBI

* Google Scholar

27. 27. Stoffel C, Dufresnes C, Okello JBA, Noirard C, Joly P, Nyakaana S, et al. Genetic consequences of population expansions and contractions in the common hippopotamus (Hippopotamus amphibius) since the Late Pleistocene. Mol Ecol. 2015;24(10):2507–20. pmid:25827243

* View Article

* PubMed/NCBI

* Google Scholar

28. 28. Thompson LG, Mosley-Thompson E, Davis ME, Henderson KA, Brecher HH, Zagorodnov VS, et al. Kilimanjaro ice core records: evidence of holocene climate change in tropical Africa. Science. 2002;298(5593):589–93. pmid:12386332

* View Article

* PubMed/NCBI

* Google Scholar

29. 29. Marchant R, Hooghiemstra H. Rapid environmental change in African and South American tropics around 4000 years before present: a review. Earth-Science Reviews. 2004;66(3–4):217–60.

* View Article

* Google Scholar

30. 30. Kiage LM, Liu K. Late Quaternary paleoenvironmental changes in East Africa: a review of multiproxy evidence from palynology, lake sediments, and associated records. Progress in Physical Geography: Earth and Environment. 2006;30(5):633–58.

* View Article

* Google Scholar

31. 31. Durant SM, Becker MS, Creel S, Bashir S, Dickman AJ, Beudels‐Jamar RC, et al. Developing fencing policies for dryland ecosystems. Journal of Applied Ecology. 2015;52(3):544–51.

* View Article

* Google Scholar

32. 32. M’soka J, Creel S, Becker MS, Murdoch JD. Ecological and anthropogenic effects on the density of migratory and resident ungulates in a human‐inhabited protected area. Afr J Ecol. 2017;55(4):618–31.

* View Article

* Google Scholar

33. 33. Watson F, Becker MS, Smit D, Droge E, Mukula T, Martens S, et al. Predation strongly limits demography of a keystone migratory herbivore in a recovering transfrontier ecosystem. Ecol Evol. 2022;12(10):e9414. pmid:36262265

* View Article

* PubMed/NCBI

* Google Scholar

34. 34. Nyirenda VR, Nkhata BA. Collaborative governance and benefit sharing in Liuwa Plain National Park, western Zambia. PARKS. 2013;19(1):103–14.

* View Article

* Google Scholar

35. 35. African Parks Network. Aerial survey confirms substantial increase in important indicator species in Liuwa. 2013.

36. 36. Dröge E, Creel S, Becker M, Christianson D, M’Soka J, Watson F. Response of wildebeest (Connochaetes taurinus) movements to spatial variation in long term risks from a complete predator guild. Biological Conservation. 2019;233139–51.

* View Article

* Google Scholar

37. 37. Dedato MN, Robert C, Taillon J, Shafer ABA, Côté SD. Demographic history and conservation genomics of caribou (Rangifer tarandus) in Québec. Evol Appl. 2022;15(12):2043–53. pmid:36540642

* View Article

* PubMed/NCBI

* Google Scholar

38. 38. Wootton E, Robert C, Taillon J, Côté SD, Shafer ABA. Genomic health is dependent on long-term population demographic history. Mol Ecol. 2023;32(8):1943–54. pmid:36704858

* View Article

* PubMed/NCBI

* Google Scholar

39. 39. Liu X, Lin L, Sinding M-HS, Bertola LD, Hanghøj K, Quinn L, et al. Introgression and disruption of migration routes have shaped the genetic integrity of wildebeest populations. Nat Commun. 2024;15(1):2921. pmid:38609362

* View Article

* PubMed/NCBI

* Google Scholar

40. 40. Allendorf FW, Hohenlohe PA, Luikart G. Genomics and the future of conservation genetics. 2010.

41. 41. Funk WC, McKay JK, Hohenlohe PA, Allendorf FW. Harnessing genomics for delineating conservation units. Trends Ecol Evol. 2012;27(9):489–96. pmid:22727017

* View Article

* PubMed/NCBI

* Google Scholar

42. 42. Corbet SW, Robinson TJ. Genetic divergence in South African Wildebeest: comparative cytogenetics and analysis of mitochondrial DNA. J Hered. 1991;82(6):447–52. pmid:1795096

* View Article

* PubMed/NCBI

* Google Scholar

43. 43. Grobler JP, Van der Bank FH. Genetic variability in South African blue wildebeest (Connochaetes oaurinus). Comp Biochem Physiol B. 1993;106(3):755–62. pmid:8281767

* View Article

* PubMed/NCBI

* Google Scholar

44. 44. Corbet SW, Grant WS, Robinson TJ. Genetic divergence in South African wildebeest: analysis of allozyme variability. J Hered. 1994;85(6):479–83. pmid:7995929

* View Article

* PubMed/NCBI

* Google Scholar

45. 45. Arctander P, Johansen C, Coutellec-Vreto MA. Phylogeography of three closely related African bovids (tribe Alcelaphini). Mol Biol Evol. 1999;16(12):1724–39. pmid:10605114

* View Article

* PubMed/NCBI

* Google Scholar

46. 46. Grobler JP, Hartl GB, Grobler N, Kotze A, Botha K, Tiedemann R. The genetic status of an isolated black wildebeest (Connochaetes gnou) population from the Abe Bailey Nature Reserve, South Africa: Microsatellite data on a putative past hybridization with blue wildebeest (C. taurinus). Mammalian Biology. 2005;70(1):35–45.

* View Article

* Google Scholar

47. 47. Excoffier L, Dupanloup I, Huerta-Sánchez E, Sousa VC, Foll M. Robust demographic inference from genomic and SNP data. PLoS Genet. 2013;9(10):e1003905. pmid:24204310

* View Article

* PubMed/NCBI

* Google Scholar

48. 48. Liu X, Fu Y-X. Exploring population size changes using SNP frequency spectra. Nat Genet. 2015;47(5):555–9. pmid:25848749

* View Article

* PubMed/NCBI

* Google Scholar

49. 49. Liu X, Fu Y-X. Stairway Plot 2: demographic history inference with folded SNP frequency spectra. Genome Biol. 2020;21(1):280. pmid:33203475

* View Article

* PubMed/NCBI

* Google Scholar

50. 50. DaCosta JM, Sorenson MD. Amplification biases and consistent recovery of loci in a double-digest RAD-seq protocol. PLoS One. 2014;9(9):e106713. pmid:25188270

* View Article

* PubMed/NCBI

* Google Scholar

51. 51. Andrews S. FastQC: a quality control tool for high throughput sequence data. In: Babraham Bioinformatics [Internet]. 2010. Available: https://www.bioinformatics.babraham.ac.uk/projects/fastqc/.

52. 52. Edgar RC. Search and clustering orders of magnitude faster than BLAST. Bioinformatics. 2010;26(19):2460–1. pmid:20709691

* View Article

* PubMed/NCBI

* Google Scholar

53. 53. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215(3):403–10. pmid:2231712

* View Article

* PubMed/NCBI

* Google Scholar

54. 54. International Sheep Genomics Consortium, Archibald AL, Cockett NE, Dalrymple BP, Faraut T, Kijas JW, et al. The sheep genome reference sequence: a work in progress. Anim Genet. 2010;41(5):449–53. pmid:20809919

* View Article

* PubMed/NCBI

* Google Scholar

55. 55. Chen L, Qiu Q, Jiang Y, Wang K, Lin Z, Li Z, et al. Large-scale ruminant genome sequencing provides insights into their evolution and distinct traits. Science. 2019;364(6446):eaav6202. pmid:31221828

* View Article

* PubMed/NCBI

* Google Scholar

56. 56. Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004;32(5):1792–7. pmid:15034147

* View Article

* PubMed/NCBI

* Google Scholar

57. 57. Kumar S, Stecher G, Li M, Knyaz C, Tamura K. MEGA X: Molecular Evolutionary Genetics Analysis across Computing Platforms. Mol Biol Evol. 2018;35(6):1547–9. pmid:29722887

* View Article

* PubMed/NCBI

* Google Scholar

58. 58. Nielsen R, Paul JS, Albrechtsen A, Song YS. Genotype and SNP calling from next-generation sequencing data. Nat Rev Genet. 2011;12(6):443–51. pmid:21587300

* View Article

* PubMed/NCBI

* Google Scholar

59. 59. Lavretsky P, Dacosta JM, Hernández-Baños BE, Engilis A Jr, Sorenson MD, Peters JL. Speciation genomics and a role for the Z chromosome in the early stages of divergence between Mexican ducks and mallards. Mol Ecol. 2015;24(21):5364–78. pmid:26414437

* View Article

* PubMed/NCBI

* Google Scholar

60. 60. DaCosta JM, Sorenson MD. ddRAD-seq phylogenetics based on nucleotide, indel, and presence-absence polymorphisms: Analyses of two avian genera with contrasting histories. Mol Phylogenet Evol. 2016;94(Pt A):122–35. pmid:26279345

* View Article

* PubMed/NCBI

* Google Scholar

61. 61. Lavretsky P, Janzen T, McCracken KG. Identifying hybrids & the genomics of hybridization: Mallards & American black ducks of Eastern North America. Ecol Evol. 2019;9(6):3470–90. pmid:30962906

* View Article

* PubMed/NCBI

* Google Scholar

62. 62. Peters JL, Lavretsky P, DaCosta JM, Bielefeld RR, Feddersen JC, Sorenson MD. Population genomic data delineate conservation units in mottled ducks (Anas fulvigula). Biological Conservation. 2016;203272–81.

* View Article

* Google Scholar

63. 63. Wickham H, Averick M, Bryan J, Chang W, McGowan L, François R, et al. Welcome to the Tidyverse. JOSS. 2019;4(43):1686.

* View Article

* Google Scholar

64. 64. R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2024. Available from: https://www.R-project.org/

65. 65. O’Leary SJ, Puritz JB, Willis SC, Hollenbeck CM, Portnoy DS. These aren’t the loci you’e looking for: Principles of effective SNP filtering for molecular ecologists. Mol Ecol. 201810.1111/mec.14792. pmid:29987880

* View Article

* PubMed/NCBI

* Google Scholar

66. 66. Emerson KJ, Merz CR, Catchen JM, Hohenlohe PA, Cresko WA, Bradshaw WE, et al. Resolving postglacial phylogeography using high-throughput sequencing. Proc Natl Acad Sci U S A. 2010;107(37):16196–200. pmid:20798348

* View Article

* PubMed/NCBI

* Google Scholar

67. 67. Willis SC, Hollenbeck CM, Puritz JB, Gold JR, Portnoy DS. Haplotyping RAD loci: an efficient method to filter paralogs and account for physical linkage. Mol Ecol Resour. 2017;17(5):955–65. pmid:28042915

* View Article

* PubMed/NCBI

* Google Scholar

68. 68. Lischer HEL, Excoffier L. PGDSpider: an automated data conversion tool for connecting population genetics and genomics programs. Bioinformatics. 2012;28(2):298–9. pmid:22110245

* View Article

* PubMed/NCBI

* Google Scholar

69. 69. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000;155(2):945–59. pmid:10835412

* View Article

* PubMed/NCBI

* Google Scholar

70. 70. Falush D, Stephens M, Pritchard JK. Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies. Genetics. 2003;164(4):1567–87. pmid:12930761

* View Article

* PubMed/NCBI

* Google Scholar

71. 71. Falush D, Stephens M, Pritchard JK. Inference of population structure using multilocus genotype data: dominant markers and null alleles. Mol Ecol Notes. 2007;7(4):574–8. pmid:18784791

* View Article

* PubMed/NCBI

* Google Scholar

72. 72. Hubisz MJ, Falush D, Stephens M, Pritchard JK. Inferring weak population structure with the assistance of sample group information. Mol Ecol Resour. 2009;9(5):1322–32. pmid:21564903

* View Article

* PubMed/NCBI

* Google Scholar

73. 73. Porras-Hurtado L, Ruiz Y, Santos C, Phillips C, Carracedo A, Lareu MV. An overview of STRUCTURE: applications, parameter settings, and supporting software. Front Genet. 2013;498. pmid:23755071

* View Article

* PubMed/NCBI

* Google Scholar

74. 74. Earl DA, vonHoldt BM. Structure Harvester: a website and program for visualizing structure output and implementing the Evanno method. Conservation Genet Resour. 2011;4(2):359–61.

* View Article

* Google Scholar

75. 75. Earl D. Structure Harvester Web v0.6.94. 2014. Available from: http://taylor0.biology.ucla.edu/structureHarvester/

* View Article

* Google Scholar

76. 76. Evanno G, Regnaut S, Goudet J. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Mol Ecol. 2005;14(8):2611–20. pmid:15969739

* View Article

* PubMed/NCBI

* Google Scholar

77. 77. Francis RM. Pophelper: an R package and web app to analyse and visualize population structure. Molecular Ecology Resources. 2017;17:27–32. https://doi.org/10.1111/1755-0998.12509

78. 78. Jombart T. adegenet: a R package for the multivariate analysis of genetic markers. Bioinformatics. 2008;24(11):1403–5. pmid:18397895

* View Article

* PubMed/NCBI

* Google Scholar

79. 79. Jombart T, Ahmed I. adegenet 1.3-1: new tools for the analysis of genome-wide SNP data. Bioinformatics. 2011;27(21):3070–1. pmid:21926124

* View Article

* PubMed/NCBI

* Google Scholar

80. 80. Paradis E. pegas: an R package for population genetics with an integrated-modular approach. Bioinformatics. 2010;26(3):419–20. pmid:20080509

* View Article

* PubMed/NCBI

* Google Scholar

81. 81. Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society Series B: Statistical Methodology. 1995;57(1):289–300.

* View Article

* Google Scholar

82. 82. Conde IC, Alvarez de JU. sgof: multiple hypothesis testing. In: R package version 2.3 [Internet]. 2016. Available: https://CRAN.R-project.org/package=sgof

* View Article

* Google Scholar

83. 83. White T, van der Ende J, Nichols TE. Beyond Bonferroni revisited: concerns over inflated false positive research findings in the fields of conservation genetics, biology, and medicine. Conserv Genet. 2019;20(4):927–37.

* View Article

* Google Scholar

84. 84. GOUDET J. hierfstat, a package for r to compute and test hierarchical F‐statistics. Molecular Ecology Notes. 2004;5(1):184–6.

* View Article

* Google Scholar

85. 85. Waples RS. A bias correction for estimates of effective population size based on linkage disequilibrium at unlinked gene loci*. Conserv Genet. 2006;7(2):167–84.

* View Article

* Google Scholar

86. 86. Waples RS, DO C. ldne: a program for estimating effective population size from data on linkage disequilibrium. Mol Ecol Resour. 2008;8(4):753–6. pmid:21585883

* View Article

* PubMed/NCBI

* Google Scholar

87. 87. Do C, Waples RS, Peel D, Macbeth GM, Tillett BJ, Ovenden JR. NeEstimator v2: re-implementation of software for the estimation of contemporary effective population size (Ne ) from genetic data. Mol Ecol Resour. 2014;14(1):209–14. pmid:23992227

* View Article

* PubMed/NCBI

* Google Scholar

88. 88. Jones AT, Ovenden JR, Wang Y-G. Improved confidence intervals for the linkage disequilibrium method for estimating effective population size. Heredity (Edinb). 2016;117(4):217–23. pmid:27005004

* View Article

* PubMed/NCBI

* Google Scholar

89. 89. Brumfield RT, Beerli P, Nickerson DA, Edwards SV. The utility of single nucleotide polymorphisms in inferences of population history. Trends in Ecology & Evolution. 2003;18(5):249–56.

* View Article

* Google Scholar

90. 90. Kumar S, Subramanian S. Mutation rates in mammalian genomes. Proc Natl Acad Sci U S A. 2002;99(2):803–8. pmid:11792858

* View Article

* PubMed/NCBI

* Google Scholar

91. 91. IUCN. Connochaetes taurinus: IUCN SSC antelope specialist group: The IUCN red list of threatened species 2008. International Union for Conservation of Nature; 2008.

* View Article

* Google Scholar

92. 92. Akaike H. A new look at the statistical model identification. IEEE Trans Automat Contr. 1974;19(6):716–23.

* View Article

* Google Scholar

93. 93. Ritz C, Spiess A-N. qpcR: an R package for sigmoidal model selection in quantitative real-time polymerase chain reaction analysis. Bioinformatics. 2008;24(13):1549–51. pmid:18482995

* View Article

* PubMed/NCBI

* Google Scholar

94. 94. Adelson DL, Raison JM, Edgar RC. Characterization and distribution of retrotransposons and simple sequence repeats in the bovine genome. Proc Natl Acad Sci U S A. 2009;106(31):12855–60. pmid:19625614

* View Article

* PubMed/NCBI

* Google Scholar

95. 95. Coetzer WG, Grobler JP. Genetic variation among different springbok (Antidorcas marsupialis) colour variants. Mammalian Biology. 2019;9942–53.

* View Article

* Google Scholar

96. 96. van Deventer R, Rhode C, Marx M, Roodt-Wilding R. The development of genome-wide single nucleotide polymorphisms in blue wildebeest using the DArTseq platform. Genomics. 2020;112(5):3455–64. pmid:32574831

* View Article

* PubMed/NCBI

* Google Scholar

97. 97. Georgiadis NJ, Kat PW, Oketch H, Patton J. Allozyme divergence within the bovidae. Evolution. 1990;44(8):2135–49. pmid:28564432

* View Article

* PubMed/NCBI

* Google Scholar

98. 98. Wright S. Evolution in Mendelian Populations. Genetics. 1931;16(2):97–159. pmid:17246615

* View Article

* PubMed/NCBI

* Google Scholar

99. 99. Frankham R. Relationship of Genetic Variation to Population Size in Wildlife. Conservation Biology. 1996;10(6):1500–8.

* View Article

* Google Scholar

100. 100. Palstra FP, Ruzzante DE. Genetic estimates of contemporary effective population size: what can they tell us about the importance of genetic stochasticity for wild population persistence?. Mol Ecol. 2008;17(15):3428–47. pmid:19160474

* View Article

* PubMed/NCBI

* Google Scholar

101. 101. O’Ryan C, Harley EH, Bruford MW, Beaumont M, Wayne RK, Cherry MI. Microsatellite analysis of genetic diversity in fragmented South African buffalo populations. Animal Conservation. 1998;1(2):85–94.

* View Article

* Google Scholar

102. 102. Palkopoulou E, Lipson M, Mallick S, Nielsen S, Rohland N, Baleka S, et al. A comprehensive genomic history of extinct and living elephants. Proc Natl Acad Sci U S A. 2018;115(11):E2566–74. pmid:29483247

* View Article

* PubMed/NCBI

* Google Scholar

103. 103. Storz JF, Beaumont MA, Alberts SC. Genetic evidence for long-term population decline in a savannah-dwelling primate: inferences from a hierarchical bayesian model. Mol Biol Evol. 2002;19(11):1981–90. pmid:12411607

* View Article

* PubMed/NCBI

* Google Scholar

104. 104. Barnosky AD, Koch PL, Feranec RS, Wing SL, Shabel AB. Assessing the causes of late Pleistocene extinctions on the continents. Science. 2004;306(5693):70–5. pmid:15459379

* View Article

* PubMed/NCBI

* Google Scholar

Citation: Szarmach SJ, Teeter KC, M’soka J, Dröge E, Ndakala H, Chifunte C, et al. (2025) Genetic diversity and demographic history of the largest remaining migratory population of brindled wildebeest (Connochaetes taurinus taurinus) in southern Africa. PLoS ONE 20(4): e0310580. https://doi.org/10.1371/journal.pone.0310580

About the Authors:

Stephanie J. Szarmach

Roles: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Visualization, Writing – original draft, Writing – review & editing

E-mail: [email protected]

Affiliations: Department of Biology, Northern Michigan University, Marquette, Michigan, United States of America, Department of Biology, Pennsylvania State University, State College, Pennsylvania, United States of America

ORICD: https://orcid.org/0000-0003-1538-8127

Katherine C. Teeter

Roles: Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing – review & editing

Affiliation: Department of Biology, Northern Michigan University, Marquette, Michigan, United States of America

Jassiel M’soka

Roles: Funding acquisition, Investigation, Writing – review & editing

Affiliation: USAID Zambia, Lusaka, Zambia

Egil Dröge

Roles: Investigation, Writing – review & editing

Affiliations: Zambian Carnivore Programme, Mfuwe, Eastern Province, Zambia, Wildlife Conservation Research Unit, Department of Biology, The Recanati-Kaplan Centre, Oxford University, Oxford, United Kingdom

Hellen Ndakala

Roles: Project administration, Resources

Affiliation: Zambia Department of National Parks and Wildlife, Liuwa Plain, Zambia

Clive Chifunte

Roles: Project administration, Resources

Affiliations: Zambian Carnivore Programme, Mfuwe, Eastern Province, Zambia, Zambia Department of National Parks and Wildlife, Mumbwa, Zambia

Matthew S. Becker

Roles: Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Writing – review & editing

Affiliations: Zambian Carnivore Programme, Mfuwe, Eastern Province, Zambia, Department of Ecology, Montana State University, Bozeman, Montana, United States of America

Alec R. Lindsay

Roles: Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing – review & editing

Affiliation: Department of Biology, Northern Michigan University, Marquette, Michigan, United States of America

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

References

1. Fryxell JM, Sinclair AR. Causes and consequences of migration by large herbivores. Trends Ecol Evol. 1988;3(9):237–41. pmid:21227239

2. Wilmshurst J, Fryxell J, Farm B, Sinclair A, Henschel C. Spatial distribution of Serengeti wildebeest in relation to resources. Journal Title. 1999;7710.

3. Boone RB, Thirgood SJ, Hopcraft JGC. Serengeti wildebeest migratory patterns modeled from rainfall and new vegetation growth. Ecology. 2006;87(8):1987–94. pmid:16937638

4. Holdo RM, Holt RD, Fryxell JM. Opposing rainfall and plant nutritional gradients best explain the wildebeest migration in the Serengeti. Am Nat. 2009;173(4):431–45. pmid:19243258

5. McNaughton SJ. Serengeti migratory wildebeest: facilitation of energy flow by grazing. Science. 1976;191(4222):92–4. pmid:17834943

6. Holdo RM, Holt RD, Coughenour MB, Ritchie ME. Plant productivity and soil nitrogen as a function of grazing, migration and fire in an African savanna. Journal of Ecology. 2006;95(1):115–28.

7. Estes RD, Raghunathan TE, van Vleck D. The Impact of Horning by Wildebeest on Woody Vegetation of the Serengeti Ecosystem. J Wildl Manage. 2008;72(7):1572.

8. Sinclair ARE. Mammal population regulation, keystone processes and ecosystem dynamics. Philos Trans R Soc Lond B Biol Sci. 2003;358(1438):1729–40. pmid:14561329

9. Hayward MW, Kerley GIH. Prey preferences of the lion (Panthera leo). Journal of Zoology. 2005;267(3):309–22.

10. Hayward MW. Prey preferences of the spotted hyaena (Crocuta crocuta) and degree of dietary overlap with the lion (Panthera leo). Journal of Zoology. 2006;270(4):606–14.

11. Creel S, Dröge E, M’soka J, Smit D, Becker M, Christianson D, et al. The relationship between direct predation and antipredator responses: a test with multiple predators and multiple prey. Ecology. 2017;98(8):2081–92. pmid:28475209

12. Dröge E, Creel S, Becker MS, M’soka J. Spatial and temporal avoidance of risk within a large carnivore guild. Ecol Evol. 2016;7(1):189–99. pmid:28070283

13. Lee DE, Kissui BM, Kiwango YA, Bond ML. Migratory herds of wildebeests and zebras indirectly affect calf survival of giraffes. Ecol Evol. 2016;6(23):8402–11. pmid:28031792

14. Estes R. The gnu’s world: Serengeti wildebeest ecology and life history. Berkeley: University of California Press; 2014.

15. Serneels S, Lambin EF. Impact of land‐use changes on the wildebeest migration in the northern part of the Serengeti–Mara ecosystem. Journal of Biogeography. 2001;28(3):391–407.

16. Homewood K, Lambin EF, Coast E, Kariuki A, Kikula I, Kivelia J, et al. Long-term changes in Serengeti-Mara wildebeest and land cover: pastoralism, population, or policies?. Proc Natl Acad Sci U S A. 2001;98(22):12544–9. pmid:11675492

17. Whyte C, Joubert I. Blue wildebeest population trends in the Kruger National Park and the effects of fencing. South African Journal of Wildlife Research. 1988;18(1):78–87.

18. Holdo RM, Fryxell JM, Sinclair ARE, Dobson A, Holt RD. Predicted impact of barriers to migration on the Serengeti wildebeest population. PLoS One. 2011;6(1):e16370. pmid:21283536

19. Mduma SAR, Sinclair ARE, Hilborn R. Food regulates the Serengeti wildebeest: a 40‐year record. Journal of Animal Ecology. 1999;68(6):1101–22.

20. Thirgood S, Mosser A, Tham S, Hopcraft G, Mwangomo E, Mlengeya T, et al. Can parks protect migratory ungulates? The case of the Serengeti wildebeest. Animal Conservation. 2004;7(2):113–20.

21. Harris G, Thirgood S, Hopcraft J, Cromsight J, Berger J. Global decline in aggregated migrations of large terrestrial mammals. Endang Species Res. 2009;755–76.

22. Heller R, Lorenzen ED, Okello JBA, Masembe C, Siegismund HR. Mid-Holocene decline in African buffalos inferred from Bayesian coalescent-based analyses of microsatellites and mitochondrial DNA. Mol Ecol. 2008;17(22):4845–58. pmid:19140976

23. Heller R, Brüniche-Olsen A, Siegismund HR. Cape buffalo mitogenomics reveals a Holocene shift in the African human-megafauna dynamics. Mol Ecol. 2012;21(16):3947–59. pmid:22725969

24. de Jager D, Glanzmann B, Möller M, Hoal E, van Helden P, Harper C, et al. High diversity, inbreeding and a dynamic Pleistocene demographic history revealed by African buffalo genomes. Sci Rep. 2021;11(1):4540. pmid:33633171

25. Okello JBA, Wittemyer G, Rasmussen HB, Arctander P, Nyakaana S, Douglas-Hamilton I, et al. Effective population size dynamics reveal impacts of historic climatic events and recent anthropogenic pressure in African elephants. Mol Ecol. 2008;17(17):3788–99. pmid:18643879

26. Okello JBA, Nyakaana S, Masembe C, Siegismund HR, Arctander P. Mitochondrial DNA variation of the common hippopotamus: evidence for a recent population expansion. Heredity (Edinb). 2005;95(3):206–15. pmid:16030528

27. Stoffel C, Dufresnes C, Okello JBA, Noirard C, Joly P, Nyakaana S, et al. Genetic consequences of population expansions and contractions in the common hippopotamus (Hippopotamus amphibius) since the Late Pleistocene. Mol Ecol. 2015;24(10):2507–20. pmid:25827243

28. Thompson LG, Mosley-Thompson E, Davis ME, Henderson KA, Brecher HH, Zagorodnov VS, et al. Kilimanjaro ice core records: evidence of holocene climate change in tropical Africa. Science. 2002;298(5593):589–93. pmid:12386332

29. Marchant R, Hooghiemstra H. Rapid environmental change in African and South American tropics around 4000 years before present: a review. Earth-Science Reviews. 2004;66(3–4):217–60.

30. Kiage LM, Liu K. Late Quaternary paleoenvironmental changes in East Africa: a review of multiproxy evidence from palynology, lake sediments, and associated records. Progress in Physical Geography: Earth and Environment. 2006;30(5):633–58.

31. Durant SM, Becker MS, Creel S, Bashir S, Dickman AJ, Beudels‐Jamar RC, et al. Developing fencing policies for dryland ecosystems. Journal of Applied Ecology. 2015;52(3):544–51.

32. M’soka J, Creel S, Becker MS, Murdoch JD. Ecological and anthropogenic effects on the density of migratory and resident ungulates in a human‐inhabited protected area. Afr J Ecol. 2017;55(4):618–31.

33. Watson F, Becker MS, Smit D, Droge E, Mukula T, Martens S, et al. Predation strongly limits demography of a keystone migratory herbivore in a recovering transfrontier ecosystem. Ecol Evol. 2022;12(10):e9414. pmid:36262265

34. Nyirenda VR, Nkhata BA. Collaborative governance and benefit sharing in Liuwa Plain National Park, western Zambia. PARKS. 2013;19(1):103–14.

35. African Parks Network. Aerial survey confirms substantial increase in important indicator species in Liuwa. 2013.

36. Dröge E, Creel S, Becker M, Christianson D, M’Soka J, Watson F. Response of wildebeest (Connochaetes taurinus) movements to spatial variation in long term risks from a complete predator guild. Biological Conservation. 2019;233139–51.

37. Dedato MN, Robert C, Taillon J, Shafer ABA, Côté SD. Demographic history and conservation genomics of caribou (Rangifer tarandus) in Québec. Evol Appl. 2022;15(12):2043–53. pmid:36540642

38. Wootton E, Robert C, Taillon J, Côté SD, Shafer ABA. Genomic health is dependent on long-term population demographic history. Mol Ecol. 2023;32(8):1943–54. pmid:36704858

39. Liu X, Lin L, Sinding M-HS, Bertola LD, Hanghøj K, Quinn L, et al. Introgression and disruption of migration routes have shaped the genetic integrity of wildebeest populations. Nat Commun. 2024;15(1):2921. pmid:38609362

40. Allendorf FW, Hohenlohe PA, Luikart G. Genomics and the future of conservation genetics. 2010.

41. Funk WC, McKay JK, Hohenlohe PA, Allendorf FW. Harnessing genomics for delineating conservation units. Trends Ecol Evol. 2012;27(9):489–96. pmid:22727017

42. Corbet SW, Robinson TJ. Genetic divergence in South African Wildebeest: comparative cytogenetics and analysis of mitochondrial DNA. J Hered. 1991;82(6):447–52. pmid:1795096

43. Grobler JP, Van der Bank FH. Genetic variability in South African blue wildebeest (Connochaetes oaurinus). Comp Biochem Physiol B. 1993;106(3):755–62. pmid:8281767

44. Corbet SW, Grant WS, Robinson TJ. Genetic divergence in South African wildebeest: analysis of allozyme variability. J Hered. 1994;85(6):479–83. pmid:7995929

45. Arctander P, Johansen C, Coutellec-Vreto MA. Phylogeography of three closely related African bovids (tribe Alcelaphini). Mol Biol Evol. 1999;16(12):1724–39. pmid:10605114

46. Grobler JP, Hartl GB, Grobler N, Kotze A, Botha K, Tiedemann R. The genetic status of an isolated black wildebeest (Connochaetes gnou) population from the Abe Bailey Nature Reserve, South Africa: Microsatellite data on a putative past hybridization with blue wildebeest (C. taurinus). Mammalian Biology. 2005;70(1):35–45.

47. Excoffier L, Dupanloup I, Huerta-Sánchez E, Sousa VC, Foll M. Robust demographic inference from genomic and SNP data. PLoS Genet. 2013;9(10):e1003905. pmid:24204310

48. Liu X, Fu Y-X. Exploring population size changes using SNP frequency spectra. Nat Genet. 2015;47(5):555–9. pmid:25848749

49. Liu X, Fu Y-X. Stairway Plot 2: demographic history inference with folded SNP frequency spectra. Genome Biol. 2020;21(1):280. pmid:33203475

50. DaCosta JM, Sorenson MD. Amplification biases and consistent recovery of loci in a double-digest RAD-seq protocol. PLoS One. 2014;9(9):e106713. pmid:25188270

51. Andrews S. FastQC: a quality control tool for high throughput sequence data. In: Babraham Bioinformatics [Internet]. 2010. Available: https://www.bioinformatics.babraham.ac.uk/projects/fastqc/.

52. Edgar RC. Search and clustering orders of magnitude faster than BLAST. Bioinformatics. 2010;26(19):2460–1. pmid:20709691

53. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215(3):403–10. pmid:2231712

54. International Sheep Genomics Consortium, Archibald AL, Cockett NE, Dalrymple BP, Faraut T, Kijas JW, et al. The sheep genome reference sequence: a work in progress. Anim Genet. 2010;41(5):449–53. pmid:20809919

55. Chen L, Qiu Q, Jiang Y, Wang K, Lin Z, Li Z, et al. Large-scale ruminant genome sequencing provides insights into their evolution and distinct traits. Science. 2019;364(6446):eaav6202. pmid:31221828

56. Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004;32(5):1792–7. pmid:15034147

57. Kumar S, Stecher G, Li M, Knyaz C, Tamura K. MEGA X: Molecular Evolutionary Genetics Analysis across Computing Platforms. Mol Biol Evol. 2018;35(6):1547–9. pmid:29722887

58. Nielsen R, Paul JS, Albrechtsen A, Song YS. Genotype and SNP calling from next-generation sequencing data. Nat Rev Genet. 2011;12(6):443–51. pmid:21587300

59. Lavretsky P, Dacosta JM, Hernández-Baños BE, Engilis A Jr, Sorenson MD, Peters JL. Speciation genomics and a role for the Z chromosome in the early stages of divergence between Mexican ducks and mallards. Mol Ecol. 2015;24(21):5364–78. pmid:26414437

60. DaCosta JM, Sorenson MD. ddRAD-seq phylogenetics based on nucleotide, indel, and presence-absence polymorphisms: Analyses of two avian genera with contrasting histories. Mol Phylogenet Evol. 2016;94(Pt A):122–35. pmid:26279345

61. Lavretsky P, Janzen T, McCracken KG. Identifying hybrids & the genomics of hybridization: Mallards & American black ducks of Eastern North America. Ecol Evol. 2019;9(6):3470–90. pmid:30962906

62. Peters JL, Lavretsky P, DaCosta JM, Bielefeld RR, Feddersen JC, Sorenson MD. Population genomic data delineate conservation units in mottled ducks (Anas fulvigula). Biological Conservation. 2016;203272–81.

63. Wickham H, Averick M, Bryan J, Chang W, McGowan L, François R, et al. Welcome to the Tidyverse. JOSS. 2019;4(43):1686.

64. R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2024. Available from: https://www.R-project.org/

65. O’Leary SJ, Puritz JB, Willis SC, Hollenbeck CM, Portnoy DS. These aren’t the loci you’e looking for: Principles of effective SNP filtering for molecular ecologists. Mol Ecol. 201810.1111/mec.14792. pmid:29987880

66. Emerson KJ, Merz CR, Catchen JM, Hohenlohe PA, Cresko WA, Bradshaw WE, et al. Resolving postglacial phylogeography using high-throughput sequencing. Proc Natl Acad Sci U S A. 2010;107(37):16196–200. pmid:20798348

67. Willis SC, Hollenbeck CM, Puritz JB, Gold JR, Portnoy DS. Haplotyping RAD loci: an efficient method to filter paralogs and account for physical linkage. Mol Ecol Resour. 2017;17(5):955–65. pmid:28042915

68. Lischer HEL, Excoffier L. PGDSpider: an automated data conversion tool for connecting population genetics and genomics programs. Bioinformatics. 2012;28(2):298–9. pmid:22110245

69. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000;155(2):945–59. pmid:10835412

70. Falush D, Stephens M, Pritchard JK. Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies. Genetics. 2003;164(4):1567–87. pmid:12930761

71. Falush D, Stephens M, Pritchard JK. Inference of population structure using multilocus genotype data: dominant markers and null alleles. Mol Ecol Notes. 2007;7(4):574–8. pmid:18784791

72. Hubisz MJ, Falush D, Stephens M, Pritchard JK. Inferring weak population structure with the assistance of sample group information. Mol Ecol Resour. 2009;9(5):1322–32. pmid:21564903

73. Porras-Hurtado L, Ruiz Y, Santos C, Phillips C, Carracedo A, Lareu MV. An overview of STRUCTURE: applications, parameter settings, and supporting software. Front Genet. 2013;498. pmid:23755071

74. Earl DA, vonHoldt BM. Structure Harvester: a website and program for visualizing structure output and implementing the Evanno method. Conservation Genet Resour. 2011;4(2):359–61.

75. Earl D. Structure Harvester Web v0.6.94. 2014. Available from: http://taylor0.biology.ucla.edu/structureHarvester/

76. Evanno G, Regnaut S, Goudet J. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Mol Ecol. 2005;14(8):2611–20. pmid:15969739

77. Francis RM. Pophelper: an R package and web app to analyse and visualize population structure. Molecular Ecology Resources. 2017;17:27–32. https://doi.org/10.1111/1755-0998.12509

78. Jombart T. adegenet: a R package for the multivariate analysis of genetic markers. Bioinformatics. 2008;24(11):1403–5. pmid:18397895

79. Jombart T, Ahmed I. adegenet 1.3-1: new tools for the analysis of genome-wide SNP data. Bioinformatics. 2011;27(21):3070–1. pmid:21926124

80. Paradis E. pegas: an R package for population genetics with an integrated-modular approach. Bioinformatics. 2010;26(3):419–20. pmid:20080509

81. Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society Series B: Statistical Methodology. 1995;57(1):289–300.

82. Conde IC, Alvarez de JU. sgof: multiple hypothesis testing. In: R package version 2.3 [Internet]. 2016. Available: https://CRAN.R-project.org/package=sgof

83. White T, van der Ende J, Nichols TE. Beyond Bonferroni revisited: concerns over inflated false positive research findings in the fields of conservation genetics, biology, and medicine. Conserv Genet. 2019;20(4):927–37.

84. GOUDET J. hierfstat, a package for r to compute and test hierarchical F‐statistics. Molecular Ecology Notes. 2004;5(1):184–6.

85. Waples RS. A bias correction for estimates of effective population size based on linkage disequilibrium at unlinked gene loci*. Conserv Genet. 2006;7(2):167–84.

86. Waples RS, DO C. ldne: a program for estimating effective population size from data on linkage disequilibrium. Mol Ecol Resour. 2008;8(4):753–6. pmid:21585883

87. Do C, Waples RS, Peel D, Macbeth GM, Tillett BJ, Ovenden JR. NeEstimator v2: re-implementation of software for the estimation of contemporary effective population size (Ne ) from genetic data. Mol Ecol Resour. 2014;14(1):209–14. pmid:23992227

88. Jones AT, Ovenden JR, Wang Y-G. Improved confidence intervals for the linkage disequilibrium method for estimating effective population size. Heredity (Edinb). 2016;117(4):217–23. pmid:27005004

89. Brumfield RT, Beerli P, Nickerson DA, Edwards SV. The utility of single nucleotide polymorphisms in inferences of population history. Trends in Ecology & Evolution. 2003;18(5):249–56.

90. Kumar S, Subramanian S. Mutation rates in mammalian genomes. Proc Natl Acad Sci U S A. 2002;99(2):803–8. pmid:11792858

91. IUCN. Connochaetes taurinus: IUCN SSC antelope specialist group: The IUCN red list of threatened species 2008. International Union for Conservation of Nature; 2008.

92. Akaike H. A new look at the statistical model identification. IEEE Trans Automat Contr. 1974;19(6):716–23.

93. Ritz C, Spiess A-N. qpcR: an R package for sigmoidal model selection in quantitative real-time polymerase chain reaction analysis. Bioinformatics. 2008;24(13):1549–51. pmid:18482995

94. Adelson DL, Raison JM, Edgar RC. Characterization and distribution of retrotransposons and simple sequence repeats in the bovine genome. Proc Natl Acad Sci U S A. 2009;106(31):12855–60. pmid:19625614

95. Coetzer WG, Grobler JP. Genetic variation among different springbok (Antidorcas marsupialis) colour variants. Mammalian Biology. 2019;9942–53.

96. van Deventer R, Rhode C, Marx M, Roodt-Wilding R. The development of genome-wide single nucleotide polymorphisms in blue wildebeest using the DArTseq platform. Genomics. 2020;112(5):3455–64. pmid:32574831

97. Georgiadis NJ, Kat PW, Oketch H, Patton J. Allozyme divergence within the bovidae. Evolution. 1990;44(8):2135–49. pmid:28564432

98. Wright S. Evolution in Mendelian Populations. Genetics. 1931;16(2):97–159. pmid:17246615

99. Frankham R. Relationship of Genetic Variation to Population Size in Wildlife. Conservation Biology. 1996;10(6):1500–8.

100. Palstra FP, Ruzzante DE. Genetic estimates of contemporary effective population size: what can they tell us about the importance of genetic stochasticity for wild population persistence?. Mol Ecol. 2008;17(15):3428–47. pmid:19160474

101. O’Ryan C, Harley EH, Bruford MW, Beaumont M, Wayne RK, Cherry MI. Microsatellite analysis of genetic diversity in fragmented South African buffalo populations. Animal Conservation. 1998;1(2):85–94.

102. Palkopoulou E, Lipson M, Mallick S, Nielsen S, Rohland N, Baleka S, et al. A comprehensive genomic history of extinct and living elephants. Proc Natl Acad Sci U S A. 2018;115(11):E2566–74. pmid:29483247

103. Storz JF, Beaumont MA, Alberts SC. Genetic evidence for long-term population decline in a savannah-dwelling primate: inferences from a hierarchical bayesian model. Mol Biol Evol. 2002;19(11):1981–90. pmid:12411607

104. Barnosky AD, Koch PL, Feranec RS, Wing SL, Shabel AB. Assessing the causes of late Pleistocene extinctions on the continents. Science. 2004;306(5693):70–5. pmid:15459379