Data from: Unraveling the mystery of the glacier bear: Genetic population structure of black bears (Ursus americanus) within the range of a rare pelage type
Data files
Dec 23, 2024 version files 84.89 KB
-
README.md
5.04 KB
-
SEAK_BB_genotypes.xlsx
79.86 KB
Abstract
Genotypes (21 microsatellite loci) of 284 black bears (Ursus americanus) from northern Southeast Alaska USA and western Canada derived from tissue samples of harvested and research animals and non-invasively collected hair samples.
Glacier bears are a rare grey color morph of American black bear (Ursus americanus) found only in northern Southeast Alaska and a small portion of western Canada. We examine contemporary genetic population structure of black bears within the geographic extent of glacier bears and explore how this structure relates to pelage color and landscape features of a recently glaciated and highly fragmented landscape. We used existing radiocollar data to quantify black bear home-range size within the geographic range of glacier bears. The mean home-range size of female black bears in the study area was 13 km2 (n = 11), whereas the home range of a single male was 86.9 km2. We genotyped 284 bears using 21 microsatellites extracted from noninvasively collected hair as well as tissue samples from harvested bears. We found ten populations of black bears in the study area, including several new populations not previously identified, divided largely by geographic features such as glaciers and marine fjords. Glacier bears were assigned to four populations found on the north and east side of Lynn Canal and the north and west side of Glacier Bay with a curious absence in the nonglaciated peninsula between. Lack of genetic relatedness and geographic continuity between black bear populations containing glacier bears suggest a possible unsampled population or an association with ice fields. Further investigation is needed to determine the genetic basis and the adaptive and evolutionary significance of the glacier bear color morph to help focus black bear conservation management to maximize and preserve genetic diversity.
README: Black bear genotypes (21 loci) from SE Alaska and Canada
https://doi.org/10.5061/dryad.p8cz8w9m1
Description of the data and file structure
Wildlife Genetics International (Nelson BC) extracted DNA with Qiagen's DNeasy tissue kits (Qiagen) and amplified and genotyped DNA as described in Paetkau et al. (1998). Sequence-based analysis of a portion of the 16S rRNA mitochondrial gene was used to confirm species (Johnson & O'Brien, 1996; Pongracz, Paetkau, Branigan, & Richardson, 2017). Individual black bears were identified using seven microsatellite markers selected for high relative variability in the study populations, including G10B, G1D, G10J, G10M, G10U, Mu50, Mu59, (Paetkau et al., 1998; Taberlet et al., 1997) plus one marker to determine sex (unpublished ZFX/ZFY primer pair designed by Wildlife Genetics International). Computerized comparisons of all pairs of unique genotypes were used to reduce genotyping errors from allelic dropout in accordance with Paetkau (2003) and Kendall et al. (2009). Microsatellite genotyping of each individual black bear was extended to include REN145PO7, G10C, CXX20, G10H, MSUT2, G10P, G1A, CPH9, CXX110, MU23, G10L, MU26, D123, and D1a (Breen et al., 2001; Kitahara, Isagi, Ishibashis, & Saitoh, 2000; Meredith, Rodzen, Banks, & Jones, 2009; Paetkau et al., 1998; Taberlet et al., 1997). Although microsatellites are less accurate than a large single-nucleotide polymorphism marker panel for detecting admixture or recent migrants (Puckett & Eggert, 2016), the panel of 21 microsatellites was highly variable (Paetkau et al., 1998) and thus informative.
Data is formatted for use in Program Structure and GenAlEx with each individual categorized into 12 regions where samples were collected.
Code/software
To assess black bear population structure, we used two individual-based Bayesian clustering approaches: STRUCTURE v2.3.3 (Pritchard, Stephens, & Donnelly, 2000) and GENELAND 3.3.0 (Guillot, Mortier, & Estoup, 2005). In STRUCTURE, we performed analyses both including and excluding the sampling location as a prior. Sampling regions were determined as clusters of sample locations separated from each other by ocean or mountains. We used an admixture model, defined allele frequencies as correlated due to suspected shared ancestry and migration between populations and performed MCMC simulations of 1,000,000 iterations with a burn-in period of 100,000. We performed 10 repetitions per simulation to test for values of K (genetic clusters) from 1 to 20. Because sample sizes were uneven across sampling locations, we determined K using the MedMeaK, MaxMeaK, MedMedK, and MaxMedK estimators (Puechmaille, 2016) using StructureSelector (Li & Liu, 2018). The maximum value of K over the 10 repetitions was determined by assigning a sampling location to a cluster if the mean or median membership coefficient ≥0.5. We calculated the number of individuals of different pelage color (black, brown, and glacier) by assigned population. We calculated population pairwise multilocus estimates of F*ST (Weir & Cockerham, 1984) using GENEPOP and Jost's *D, which measures genetic differentiation using differences in true allelic diversity correcting for sampling bias, (Jost, 2008) using GenAlEx 6.5 (Peakall & Smouse, 2006). Ten thousand random permutations were used to test significance of pairwise F*ST and Jost's *D.
Methods
From 2002–2014, we collected black bear tissue samples from harvested bears and bears captured for research purposes, as well as non-invasive hair samples from rub trees, hair snares, and scented hair traps from multiple research projects in 12 sampling regions.
Wildlife Genetics International (Nelson BC, Canada) extracted DNA with Qiagen’s DNeasy tissue kits (Qiagen, Toronto Canada) and amplified and genotyped DNA as described in Paetkau et al. (1998). Sequence-based analysis of a portion of the 16S rRNA mitochondrial gene was used to confirm species (Johnson and O’Brien 1996). Individual black bears were identified using eight highly variable microsatellite loci including G10B, G1D, G10J, G10M, G10U, Mu50, Mu59, (Taberlet et al. 1997, Paetkau et al. 1998) and sex (unpublished ZFX/ZFY primer pair designed by Wildlife Genetics International). Computerized comparisons of all pairs of unique genotypes were used to reduce genotyping errors from allelic dropout in accordance with Paetkau (2003) and Kendall et al. (2009). Microsatellite genotyping of each individual black bear was extended to include REN145PO7, G10C, CXX20, G10H, MSUT2, G10P, G1A, CPH9, CXX110, MU23, G10L, MU26, D123, and D1a (Taberlet et al. 1997, Paetkau et al. 1998, Kitahara et al. 2000, Breen et al. 2001, and Meredith et al. 2009).
Usage notes
The data is formatted to be compatible with programs GenAlEx and Structure