Figures
Abstract
Recent data indicates that blotched tiger salamanders (Ambystoma tigrinum melanostictum) in northern regions of Yellowstone National Park are declining due to climate-related habitat changes. In this study, we used ancient and modern mitochondrial haplotype diversity to model the effective size of this amphibian population through recent geological time and to assess past responses to climatic changes in the region. Using subfossils collected from a cave in northern Yellowstone, we analyzed >700 base pairs of mitochondrial sequence from 16 samples ranging in age from 100 to 3300 years old and found that all shared an identical haplotype. Although mitochondrial diversity was extremely low within the living population, we still were able to detect geographic subdivision within the local area. Using serial coalescent modelling with Bayesian priors from both modern and ancient genetic data we simulated a range of probable population sizes and mutation rates through time. Our simulations suggest that regional mitochondrial diversity has remained relatively constant even through climatic fluctuations of recent millennia.
Citation: McMenamin SK, Hadly EA (2012) Ancient DNA Assessment of Tiger Salamander Population in Yellowstone National Park. PLoS ONE 7(3): e32763. https://doi.org/10.1371/journal.pone.0032763
Editor: David Nogues-Bravo, University of Copenhagen, Denmark
Received: October 20, 2011; Accepted: February 2, 2012; Published: March 12, 2012
Copyright: © 2012 McMenamin, Hadly. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was funded by the Stanford Center for Evolutionary Studies (no grant number or URL; contact: Ward Watt wbwatt@stanford.edu) and by two Grant-in-Aid of Research grants from the Sigma Xi Society (grant numbers G20071091712537615 and G20061041426504729; http://www.sigmaxi.org/). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Detailed contemporary genetic characteristics of a population can be analyzed to reveal aspects of its history. Nonetheless, the modern genetics of a population serve only as a snapshot of complex and dynamic processes averaged over time and biased by recent events. Incorporating ancient information into genetic studies can place a population into its historical context, and can reveal details of a population's history that might otherwise be lost with time [1], [2]. Especially when considering a population facing contemporary environmental threats, it can be highly informative to determine how the population responded to comparable historic environmental changes [3].
Amphibian populations are sensitive environmental indicators, tracking both aquatic and terrestrial environments, and their responses to historical environmental changes may lend insight into their prospects in rapidly changing modern ecosystems. Genetic variation in modern populations of tiger salamanders (Ambystoma spp.) suggest complex colonization, migration and bottleneck patterns through their histories that could be explored in greater detail by anchoring modern genetic variation in ancient variation. Nonetheless, no previous study has included assessments of ancient amphibian DNA, either in a phylogeographic or a population genetic context. To our knowledge, the oldest analyzed amphibian genetic material was extracted from museum specimens accessioned in 1920 [4]. In the current study, we used amphibian samples from the last 3000 years to study the history of an amphibian population that has likely experienced declines and bottlenecks in past millennia and centuries [5], as well as in recent decades [6].
Amphibians may be declining in northern Yellowstone. Letters from the late 1800s and reports as recent as the 1950s describe summer mass migrations of hundreds of thousands of adult A. t. melanostictum [7], but recent data suggest the population may be in decline due to changing climate, pond desiccation [6] and disease. The genetics of Yellowstone Ambystoma further exhibit signatures of population decline and bottlenecks: microsatellites show a deficit of rare alleles, an excess of heterozygosity and a notably ragged distribution of allele sizes [5]. However, the ragged distribution of allele sizes likely reflect bottleneck events older than the recent decline trends, and may reflect the initial colonization bottleneck after deglaciation ∼14,000 years before present (ybp). More recently, population bottlenecks may have been caused by specific climatic events such as the Medieval Warm Period (∼1200 ybp), long-term climatic fluctuations or by the historic practice of stocking ponds with fish that prey on salamanders [8]. More powerful methods can be used to date bottlenecks from modern genetic information. However in the absence of ancient genetic data it can be difficult to pinpoint with great precision when any particular bottleneck occurred and to which factors it may be attributed.
Lamar Cave in northern Yellowstone National Park, Wyoming, USA, is a rich wood rat (Neotoma cinerea) midden deposit chronicling over 3000 years of vertebrate community history [9]. Generations of wood rats collected and accumulated raptor pellets and carnivore scats containing vertebrate specimens originating within 5–7 km of the site [10], and this natural collection has been used to describe genetic, phenotypic and compositional changes in mammalian populations through time [11], [12], [13]. The blotched tiger salamander Ambystoma tigrinum melanostictum is the most abundant non-mammalian vertebrate present in the deposit, and Bruzgul et al. [14] used cervical vertebrae to characterize the demographics of the amphibian population through time, concluding that growth and development tracked contemporary climatic conditions over thousands of years.
In this study, we examined the mitochondrial genetics of this population across several localities in the local area and through recent geologic time, using serial coalescent modeling to develop a distribution of population sizes through time. Using parsimony based on our population genetic data, we aimed to disentangle the relative severities of ancient versus recent population declines and to determine the size at which this population has been maintained through environmental events. We focused on a region of the mitochondrial genome spanning the tRNA gene for threonine, a highly polymorphic non-coding intergenic spacer region, the tRNA gene for proline, and the D-loop region. The D-loop region is frequently targeted in studies using aDNA (see [15], [16]).
Materials and Methods
Ethics Statement
Handling of animals was conducted according to institutional and national guidelines. All work reviewed by the Stanford University Administrative Panel on Laboratory Animal Care and all protocols were approved under permit number 10284. All necessary approval was obtained for the described field studies through the Yellowstone National Park Research Permit Office under permit number YELL-2008-SCI-5638.
Ancient sample selection and amplification
Specimens of A. t. melanostictum were collected and sorted during excavations of Lamar Cave, a paleontological site in Yellowstone National Park (Fig. 1). The specimens were dated by stratigraphic layer, and layer age was determined through radiocarbon dating of 18 representative samples from the 16 different layers present [9]. Fifteen cervical A. tigrinum vertebra samples were chosen from 10 different stratigraphic layers, ranging in age from 100 to 3000 years old (Table 1). Each individual has a single cervical vertebra and the use of this bone ensured that each individual could be represented in our dataset only once.
Map area shows the Lamar Valley in northern Yellowstone National Park in Wyoming, USA. Thin grey lines show rivers, thick grey lines show park and state boundaries, black lines show roads. Lamar Cave site indicated with arrow. Samples collected from Lamar Cave = 16 (all Haplotype 1), samples collected from Ice Lake (Ice) = 17, Rainbow Lake (R) = 12, Pond A = 7, Pond F = 12, Pond J = 16, Pond 47 = 5, Pond 1 = 5, Pond 5 = 11, Pond 26 = 18, Pond 34 = 8, Pond 38 = 6, Pond 36 = 17, Pond 49 = 9.
We adhered to strict extraction and amplification protocols, and numerous precautions were taken to prevent contamination (see [12], [13], [17]). Neither personnel nor facilities had contact with any amphibian material prior to processing the ancient samples, and all field and lab work with modern A. t. melanostictum samples was initiated only after all ancient samples had been processed and analyzed. Ancient DNA (aDNA) was extracted from samples in a designated aDNA facility physically isolated from any area in which amplification or analyses were performed. Personnel were not permitted to enter the aDNA area for 24 h after contact with PCR products or facilities. Prior to extracting aDNA, all the surfaces in the aDNA facility were cleaned with 25% bleach. The room, the inside of the flow hood, and all reagents and materials were irradiated with UV light for a minimum of 12 hours. Extractions and PCR setup were performed in a sterilized Class II laminar flow hood within the sterilized aDNA facility.
Extractions of ancient material were performed in small batches of 5 or fewer samples. Weight of sampled cervical vertebrae was measured and ranged from 1 to 12 mg (average = 5.4 mg). After crushing the vertebra inside of sterilized aluminum foil using a sterilized mortar and pestle, powdered material was digested in 275 µl 0.5 M EDTA pH 8.0 with 0.5% sodium dodecyl sulfate and 100 µg/ml proteinase K for 24 h at 55°C, then purified using Qiaquick® DNA purification silica columns (Qiagen Inc., Valencia, California). We included two dummy negative controls in this process, the first control added with the samples as they were weighed and powdered, and the second added at the addition of buffer during the extraction phase. Each of these controls was used alongside the samples through the amplification process to monitor for cross-contamination. A sample of the sequences were cloned and sequenced to verify sequence fidelity (see [18]).
We designed five sets of overlapping primers, altogether 729 base pairs (bp) of the mitochondrial genome:
1F: AAACATCGATCTTGTAAGTCG 1R: ATTAAAAAAGCTCCTACGCTT (to amplify a 184 bp product); 2F: CTTAAAATCAGTAAACTGC 2R: CAAGAGGGGAGGATTTTCAC (201 bp product); 3F: CAACTCGAATTTTCTATTGC 3R: CATCCAACTATCTGCCACACG (214 bp product); 4F GTAACGTGGCAACATATTATG 4R: CTGGTTAAAATCTATGGAC (201 bp product); 5F: GAGTGCCTTACTTCCCTTG 5R: GTAATATGACTGGTACTAC (205 bp product).
Amplifications were performed in 50 µL reactions containing 1 U AmpliTaq® Gold polymerase (Applied Biosystems, Foster City, California), 1× Taq Gold buffer, 5 mM MgCl2, 1 mM each dNTP, 1.3 mg/ml spermidine (Sigma-Aldrich, St. Louis, Missouri) and 0.25 µM of each primer. A third control was added during PCR preparation.
Samples were transferred into the physically isolated PCR room and run on a 9700 thermocycler (Applied Biosystems, Foster City, California) for 10 min at 95°C, 45 cycles of 95°C for 30 sec, 45°C for 30 sec, 72°C for 30 sec, followed by 10 min at 72°C. Every amplification reaction included a negative control with no added DNA. We tested each of the dummy negative controls collected during the extraction process, of which there were eight in total (two for each of the four extraction batches performed). These were amplified along with the actual DNA samples, and none of the dummy negative controls ever yielded product visible on an agarose gel stained with ethidium bromide. Likewise, the negative control amplification reactions containing no DNA never yielded detectable products. Our PCRs were successful for 51% of positive (DNA-containing) reactions. We found no correlation between amplification success and fragment size or sample age. Amplified fragments were purified with the Qiaquick® PCR purification kit and sequenced in both forward and reverse directions by Genaissance Pharmaceuticals (New Haven, Connecticut; now Cogenics, Inc., Houston, Texas).
Modern sample collection and amplification
Collection and analysis of modern individuals was performed only after all ancient analyses were completed. Live A. t. melanostictum were collected from 13 ponds in northern Yellowstone National Park (Fig. 1) using either dip nets or funnel traps, and between 8 and 18 individuals were collected and analyzed from each location (average = 12.4). Tail clips of 3–10 mm were collected from each individual and stored in vials containing 90–100% EtOH and refrigerated until extraction; live specimens were then released where captured. DNA was extracted using either a DNeasy® Blood and Tissue kit (Qiagen Inc., Valencia, California) or by incubating tissue with 250 µl 5% Chelex 100 beads (Sigma-Aldrich, St. Louis, Missouri) for 95°C for 1 h. Amplifications were performed in 50 µL reactions containing 1 U Taq polymerase (Invitrogen, Carlsbad, California), 1× PCR buffer, 2 mM MgCl2, 0.2 mM each dNTP and 0.25 µM of primers 1F and 5R (see sequences in previous section). PCRs were run at 95°C for 2 min, 35 cycles of 95°C for 30 sec, 55°C for 30 sec, 72°C for 1 min, followed by 5 min at 72°C. PCR samples were sequenced in both forward and reverse directions by Cogenics, Inc. (Houston, Texas, formerly Genaissance Pharmaceuticals). In order authenticate our sequences (see [18]), we randomly selected seven samples and cloned each into PCR-TOPO TA vectors according to the manufacturers protocol (Invitrogen, Carlsbad, California). Ten colonies were sequenced for each sample with M13 primers, and we verified that each corresponded to the original sequence.
Population Analyses and Coalescent Modeling
Sequence reads were aligned with Sequencher software (version 3.1.1, Gene Codes Corporation, Ann Arbor, Michigan). Mantel tests and AMOVA, both with 9999 permutations, were performed for the modern population using GenAlEx 6.2 [19], [20]. GenAlEx was also used to calculate φPT, a measure of genetic variability analogous to FST. Tajima's D was calculated in Arlequin 3.0 [21]. We used Bayesian Serial SimCoal [22] to model the most likely effective population sizes by simulating haplotype frequencies through time. Bayesian Serial SimCoal is similar to SIMCOAL 1.0 [23], but allows the input of Bayesian priors into the models and incorporate both modern and ancient sampling into simulations. Several mitochondrial mutation rates have been proposed for Ambystoma [24], [25] and rates of substitution vary between different regions of the mitochondrial genome; we therefore explored a range of mutation rates, from 0.25% to 1.75% mutations per nucleotide per million years. We assumed a generation time of four years. Samples from Lamar Cave are known to have originated within a 7-km radius of the cave site [10]. Therefore, we modeled ponds in the Lamar Valley (Fig. 1 inset), all within 7 km of the cave, as the modern descendants of the populations represented in Lamar Cave. Using ranges of priors for population size (0–40000) and mutation rate (0.25%–1.75%), we calculated the rate at which simulations produced outcomes matching the observed number of ancient mitochondrial segregating sites (i.e. 0 sites) or both the observed ancient segregating sites (0 sites) as well as the number of modern segregating sites for the Lamar Valley population (i.e. 1 site). After running Bayesian models, we ultimately performed simulations based on fixed parameters. In these simulations, we fixed both the mutation rate and the population size, then ran 1000 independent simulations and evaluated the percentage of successful simulations. A simulation was deemed successful when it produced the appropriate numbers of segregating sites in the ancient (Ancient Data) or in both the modern and ancient populations (Ancient+Modern Data). We calculated the probability of any potential population size as the percent success in these simulations.
Results
We sampled 143 modern and 16 ancient A. t. melanostictum from northern Yellowstone (Fig. 1) and amplified more than 700 bp of the mitochondrial genome spanning both coding and non-coding regions. We identified only four unique haplotypes (GenBank Accession Numbers: GQ260669–GQ260817 and GQ260834–GQ260845). Each haplotype differed from the others by a single non-coding polymorphic site. The polymorphisms distinguishing Haplotypes 2 and 3 were in the D-loop, and the haplotype distinguishing Haplotype 4 was in the intergenic spacer region. Haplotype 1 (H1) was by far the most common genotype, present at every location and representing 83% of modern samples. H1 dominated the Lamar Valley region, where five out of seven sampled pond populations were composed exclusively of H1 (Fig. 1). All 16 of the ancient individuals genotyped also represented H1.
Multiple AMOVAs of the modern genetic data indicated that the population exhibited significant isolation by distance (Table 2). A Mantel test showed Rxy = 0.23 (p<0.001) with distance explaining 5.1% variation between ponds. The amount of subdivision within the population was significant (φPT = 0.179, p = 0.001). We calculated Tajima's D for individual ponds, for the ponds in Lamar Valley and for all ponds; in all scenarios, there was no significant deviation from 0.
We used Bayesian serial coalescent modeling [22], [23] to explore probable effective population sizes. Considering the Lamar Valley (which contained two haplotypes) and the ancient data (which contained a single haplotype), we determined that Ne was likely to be between 5000 and 10000 individuals (Fig. 2), and that the population has remained at a relatively constant size through time.
A: Results of simulating the number of segregating sites observed in ancient samples from Lamar Cave. B: Results of simulating segregating site frequency found in modern Lamar Valley. Mutation rates represent rate of mutation per base pair per million years.
Discussion
The low mitochondrial diversity that we observed in modern A. t. melanostictum is comparable to the low mtDNA diversity described in other Ambystoma populations [26], [27], and consistent with the low nuclear microsatellite diversity described for this population [28].
H1 shares more than 99% identity with mitochondrial sequences of A. t. melanostictum collected from other geographic regions in the Great Plains and the Pacific Northwest [24]. All of the Yellowstone haplotypes showed the greatest similarity to A. t. melanostictum collected from the eastern Great Plains. The greater Yellowstone ecosystem was covered by ice until ∼14,000 ybp, and the region was colonized by amphibians soon after this ice cap melted, and this genetic similarity is consistent with a recolonization westward from the Great Plains. The low diversity throughout the region suggests that there has been relatively little gene exchange with outside populations since this initial colonization, however the mutation rate and overall diversity is so low that moderate levels of migration and gene exchange cannot be excluded, especially from populations with identical haplotypes present. Although diversity was low, the distribution of the four haplotypes revealed significant spatial differentiation (Fig. 1). We found significant isolation-by-distance as well as significant population subdivision (Table 2). The amount of spatial partitioning even with such low genetic diversity suggests that migration between ponds is low and that philopatry is high.
Our analysis of aDNA indicated that H1 was historically prevalent in this population as well; indeed H1 was the only haplotype identified in the ancient samples. Rodríguez and colleagues [29] recently found a similar lack of mitochondrial variation in 19 ancient specimens of the Iberian linx. Based on previous analyses of strontium isotopes, Lamar Cave specimens are known to originate within a 7-km radius of the cave site [10]. We therefore used the ponds in Lamar Valley (Fig. 1 inset), all within 7 km of the cave, as the modern descendants of the population represented in Lamar Cave. Within the modern Lamar Valley population, 71 of 74 (96%) samples represented H1. If haplotype frequencies had remained entirely consistent over the past several millennia, the chance of independently sampling H1 in all of the 16 samples would be 0.9616 = 0.52 or 52%. This relatively large probability suggests that the spatial distribution of modern haplotypes reflects patterns that have likely persisted for millennia, and that the haplotype distribution seen in the current Lamar Valley is similar to the distribution that has existed for several thousand years.
Of course, haplotype frequencies do change through time and even if populations remain geographically stable, new haplotypes are introduced by mutation. To construct models representing the dynamic nature of population genetics, we simulated coalescence based on a range of population sizes and mutation rates [22]. The mitochondrial mutation rate in Ambystoma may be as high as 1.5% mutations per base pair per million years in the D-loop, but is probably closer to 0.75% in coding regions such as the cytochrome-b gene [24], [30]. The region sequenced spans coding regions, an intergenic spacer region and the D-loop, the mutation rate is certainly not consistent across all sequenced base pairs. Indeed, we did not observe any polymorphisms in any of the sequenced coding region. We used a relaxed Bayesian molecular clock to explore a range of mutation rates, and present the upper (1.75%) and lower (0.25%) expected limits of mutation rates. We calculated Tajima's D for the Lamar Valley population (as well for the entire population) to be indistinguishable from zero, and we thus used a constant population size for all of our models. The effective population size of Lamar Valley A. t. melanostictum was simulated using the number of mitochondrial segregating sites identified in ancient (Fig. 2A) or identified in both modern and ancient (Fig. 2B) to evaluate the success of each simulation. When we considered only the ancient individuals and segregating sites, we found a small population to be maximally likely (Fig. 2A). Considering both modern and ancient individuals and segregating sites, we estimated the effective population size to be between 3000 and 10000 (Fig. 2B). We describe a population that has experienced little genetic change through time and shows no variation in aDNA, reminiscent of the situation seen in some mammalian populations [29].
Base pair misincorporations, specifically cytosine-deamination transitions are shown to be pervasive in amplifications of ancient and historic DNA, even from samples as young as 50 years [31], [32], [33] showed that base pair misincorporations, specifically cytosine deamination-mediated transitions, are pervasive in amplifications. However, in this study, we genotyped 16 A. t. melanostictum ranging in age from 100 to 3300 years, sequencing a total of more than 10,000 ancient base pairs with a minimum of 2× coverage. PCR misincorporations were not detected in any of the sequences, showing that DNA from all these samples could be faithfully amplified. While our laboratory has performed numerous aDNA-based studies (i.e. [12], [13], [17], [34], [35]), all these studies were based on mammalian DNA, and amphibian material had never been handled in our laboratory. We performed all the work with ancient amphibian DNA before any of the modern DNA or tissue entered our laboratory, and indeed before the field work was performed, making it highly unlikely that the ancient haplotype we isolated was the result of contamination with modern genetic material.
The vertebrate specimens used in the present study, like other samples from Lamar Cave [12], [13], [15], [36] were consumed by raptors or mammal predators, regurgitated or defecated, and then aged in cave conditions [9], [37], [38]. In contrast, ancient DNA that shows frequent cytosine deamination (in studies such as [39]) originates from bones and teeth of large mammals (wolf, mammoth, bear). The specimens shown to be prone to deamination were not digested as were our samples, but rather decomposed, probably in exposed environments. It is possible that digestion and cave preservation in cool temperate regions are less detrimental to genetic fidelity than the process of exposed decomposition. DNA sequences extracted from late-Holocene specimens often contain lesions that result in misincorporations during PCR. However, this fact should not be extrapolated to mean that every sequence derived from material older than 50 years is unreliable. The conditions under which a sample is preserved appear to play a key role in DNA viability, and sequencing coverage as well as the physical history of a specimen (i.e. its taphonomy) should be considered in assessing credibility of ancient sequence. Indeed, studies of Neanderthal DNA indicate that the temperature and conditions under which samples aged is more important to DNA preservation than age [40], [41].
Signals of population bottlenecks are evident across northern Yellowstone, and these have been attributed to the bottlenecks induced by recent events such as fish stocking in the 1950s [5]. Observing only the modern lack of mitochondrial diversity in the lower Lamar Valley, combined with the genetic signals of bottlenecking and modern evidence of population decline, we would reasonably assume that diversity had recently been lost through a discrete population decline event. Using subfossils from the same area, we describe a population that has undergone little to no genetic change over the last several thousand years. Thus, the additional perspective of ancient genetics suggests that this population has been maintained at a relatively constant size for several millennia. Though they may indeed have been detrimental to local A. t. melanostictum population sizes, we conclude that the bottlenecks caused by fish stocking and disease are not the primary cause of the low mitochondrial diversity in this region.
Acknowledgments
W. Love-Anderegg and L. Palumbi, Y. J. Lee, A. T. Eamsherangkoon, C. L. Li and particularly L. Ha assisted greatly in the field and in the lab. Many thanks to Z. Zhen and C. Anderson for technical assistance and support, and to J. L. Blois, R. Adams, T. Root, S. Palumbi, S. R. Voss and W. Watt for discussion and comments. Park staff, particularly C. Hendrix, C. Smith, C. Daigle-Berg and S. Gunther facilitated fieldwork. Ancient specimens used in this project were sorted and archived by W. Long.
Author Contributions
Conceived and designed the experiments: SM EH. Performed the experiments: SM. Analyzed the data: SM. Contributed reagents/materials/analysis tools: SM EH. Wrote the paper: SM EH.
References
- 1. Ramakrishnan U, Hadly E (2009) Using phylochronology to reveal cryptic population histories: review and synthesis of 29 ancient DNA studies. Molecular Ecology 18: 1310–1330.
- 2. Lorenzen ED, Nogués-Bravo D, Orlando L, Weinstock J, Binladen J, et al. (2011) Species-specific responses of Late Quaternary megafauna to climate and humans. Nature 479: 359–364.
- 3. de Bruyn M, Hoelzel AR, Carvalho GR, Hofreiter M (2011) Faunal histories from Holocene ancient DNA. Trends in Ecology & Evolution 26: 405–413.
- 4. Carranza S, Arnold E (2003) History of West Mediterranean newts, Pleurodeles (Amphibia: Salamandridae), inferred from old and recent DNA sequences. Systematics and Biodiversity 1: 327–337.
- 5. Spear SF, Peterson CR, Matocq MD, Storfer A (2006) Molecular evidence for historical and recent population size reductions of tiger salamanders (Ambystoma tigrinum) in Yellowstone National Park. Conservation Genetics 7: 605–611.
- 6. McMenamin SK, Hadly EA, Wright C (2008) Climatic change and wetland desiccation cause amphibian decline in Yellowstone National Park. Proceedings of the National Academy of Sciences, USA 105: 16988–16993.
- 7.
Koch ED, Peterson CR (1995) Amphibians & Reptiles of Yellowstone and Grand Teton National Parks. Salt Lake City: University of Utah Press. 188 p.
- 8.
Varley JD (1981) A history of fish stocking activites in Yellowstone National Park between 1881–1980. Yellowstone: National Park Service.
- 9. Hadly EA (1996) Influence of Late-Holocene Climate on Northern Rocky Mountain Mammals. Quaterary Research 46: 298–310.
- 10. Porder S, Paytan A, Hadly EA (2003) Mapping the origin of faunal assemblages using Strontium isotopes. Paleobiology 29: 197–204.
- 11. Hadly EA, Maurer BA (2001) Spatial and temporal patterns of species diversity in montane mammal communities of western North America. Evolutionary Ecology Research 3: 477–486.
- 12. Hadly EA, Ramakrishnan U, Chan YL, van Tuinen M, O'Keefe K, et al. (2004) Genetic Response to Climatic Change: Insights from Ancient DNA and Phylochronology. PLoS Biology 2: e290.
- 13. Hadly EA, Kohn MH, Leonard JA, Wayne RK (1998) A genetic record of population isolation in pocket gophers during Holocene climatic change. Proceedings of the National Academy of Sciences, USA 95: 6893–6896.
- 14. Bruzgul JE, Long W, Hadly EA (2005) Temporal response of the tiger salamander (Ambystoma tigrinum) to 3,000 years of climatic variation. BioMed Central Ecology 5: 7.
- 15. Van Tuinen M, O'Keefe KIM, Ramakrishnan UMA, Hadly EA (2008) Fire and ice: genetic structure of the Uinta ground squirrel (Spermophilus armatus) across the Yellowstone hotspot. Molecular Ecology 17: 1776–1788.
- 16. Yang DY, Speller CF (2006) Co-amplification of cytochrome b and D-loop mtDNA fragments for the identification of degraded DNA samples. Molecular Ecology Notes 6: 605–608.
- 17. Hadly EA, Van Tuinen M, Chan Y, Heiman K (2003) Ancient DNA evidence of prolonged population persistence with negligible genetic diversity in endemic tuco-tuco (Ctenomys sociabilis). Journal of Mammalogy 84: 403–417.
- 18. Winters M, Barta JL, Monroe C, Kemp BM (2011) To Clone or Not To Clone: Method Analysis for Retrieving Consensus Sequences In Ancient DNA Samples. PloS One 6: e21247.
- 19. Peakall R, Smouse PE (2006) GENALEX 6: genetic analyses in Excel. Population genetic software for teaching and research. Molecular Ecology Notes 6: 288–295.
- 20. Smouse PE, Peakall R, Gonzales E (2008) A heterogeneity test for fine-scale genetic structure. Molecular Ecology 17: 3389–3400.
- 21. Excoffier L, Laval G, Schneider S (2005) Arlequin version 3.0: an integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online 1: 47–50.
- 22. Anderson CNK, Ramakrishnan U, Chan YL, Hadly EA (2005) Serial SimCoal: A population genetics model for data from multiple populations and points in time. Bioinformatics 21: 1733–1734.
- 23. Excoffier L, Novembre J, Schneider S (2000) SIMCOAL: a general coalescent program for simulation of molecular data in interconnected populations with arbitrary demography. Journal of Heredity 91: 506–509.
- 24. Shaffer HB, McKnight ML (1996) The Polytypic Species Revisited: Genetic Differentiation and Molecular Phylogenetics of the Tiger Salamander Ambystoma tigrinum (Amphibia: Caudata) Complex. Evolution 50: 417–433.
- 25. Bos D, Gopurenko D, Williams R, DeWoody A, Knowles L (2008) Inferring population history and demography using microsatellites, mitochondrial DNA, and major histocompatibility complex (MHC) genes. Evolution 62: 1458–1468.
- 26. Templeton AR, Routman E, Phillips CA (1995) Separating Population Structure from Population History: A Cladistic Analysis of the Geographical Distribution of Mitochondrial DNA Haplotypes in the Tiger Salamander, Ambystoma tigrinum. Genetics 140: 767–782.
- 27. Samuels AK, Weisrock DW, Smith JJ, France KJ, Walker JA, Putta S, Voss SR (2005) Transcriptional and phylogenetic analysis of five complete ambystomatid salamander mitochondrial genomes. Gene 349: 43–53.
- 28. Spear SF, Peterson CR, Matocq MD, Storfer A (2005) Landscape genetics of the blotched tiger salamander (Ambystoma tigrinum melanostictum). Molecular Ecology 14: 2553–2564.
- 29. Rodríguez R, Ramírez O, Valdiosera C, García N, Alda F, et al. (2011) 50,000 years of genetic uniformity in the critically endangered Iberian lynx. Molecular Ecology 20: 3785–3795.
- 30. Church SA, Kraus JM, Mitchell JC, Church DR, Taylor DR (2003) Evicence for multiple Pleistocene refugia in the postglacial expansion of the Eastern Tiger Salamander Ambystoma tigrinum tigrinum. Evolution 57: 372–383.
- 31. Brotherton P, Endicott P, Sanchez JJ, Beaumont M, Barnett R, et al. (2007) Novel high-resolution characterization of ancient DNA reveals C>U-type base modification events as the sole cause of post mortem miscoding lesions. Nucleic Acids Research 35: 5717–5728.
- 32.
Fulton TL (2012) Setting up an ancient DNA laboratory; Hofreiter BSaM, editor. Clifton, NJ: Springer. 1 p.
- 33.
Stiller M, Green RE, Ronan M, Simons JF, Du L, et al. (2006) Patterns of nucleotide misincorporations during enzymatic amplification and direct large-scale sequencing of ancient DNA. Proceedings of the National Academy of Sciences, USA.
- 34. Chan Y, Lacey E, Pearson O, Hadly E (2005) Ancient DNA reveals Holocene loss of genetic diversity in a South American rodent. Biology Letters 1: 423–426.
- 35. Chan Y, Anderson C, Hadly E (2006) Bayesian Estimation of the Timing and Severity of a Population Bottleneck from Ancient DNA. PLoS Genetics 2: e59.
- 36. Spaeth P, Van Tuinen M, Chan Y, Terca D, Hadly E (2009) Phylogeography of Microtus longicaudus in the tectonically and glacially dynamic central Rocky Mountains. Journal of Mammalogy 90: 571–584.
- 37. Hadly EA (1999) Fidelity of terrestrial vertebrate fossils to a modern ecosystem. Palaeogeography, Palaeoclimatology, Palaeoecology 149: 389–409.
- 38. Hadly EA (1997) Evolutionary and ecological response of pocket gophers (Thomomys talpoides) to late-Holocene climatic change. Biological Journal of the Linnean Society 60: 277–296.
- 39. Hofreiter M, Jaenicke V, Serre D, von Haeseler A, Pääbo S (2001) DNA sequences from multiple amplifications reveal artifacts induced by cytosine deamination in ancient DNA. Nucleic Acids Research 29: 4793–4799.
- 40. Smith CI, Chamberlain AT, Riley MS, Cooper A, Stringer CB, et al. (2001) Neanderthal DNA: Not just old but old and cold? Nature 410: 771–772.
- 41. Smith CI, Chamberlain AT, Riley MS, Stringer C, Collins MJ (2003) The thermal history of human fossils and the likelihood of successful DNA amplification. Journal of Human Evolution 45: 203–217.