Utilising a reconstructed ancestral mitochondrial genome of a clade to design hybridisation capture baits can provide the opportunity for recovering mitochondrial sequences from all its descendent and even sister lineages. This approach is useful for taxa with no extant close relatives, as is often the case for rare or extinct species, and is a viable approach for the analysis of historical museum specimens. Asiatic linsangs (genus Prionodon) exemplify this situation, being rare Southeast Asian carnivores for which little molecular data is available. Using ancestral capture we recover partial mitochondrial genome sequences for seven banded linsangs (P. linsang) from historical specimens, representing the first intraspecific genetic dataset for this species. We additionally assemble a high quality mitogenome for the banded linsang using shotgun sequencing for time-calibrated phylogenetic analysis. This reveals a deep divergence between the two Asiatic linsang species (P. linsang, P. pardicolor), with an estimated divergence of ~12 million years (Ma). Although our sample size precludes any robust interpretation of the population structure of the banded linsang, we recover two distinct matrilines with an estimated tMRCA of ~1 Ma. Our results can be used as a basis for further investigation of the Asiatic linsangs, and further demonstrate the utility of ancestral capture for studying divergent taxa without close relatives.
Citation: Paijmans JLA, Barlow A, Henneberger K, Fickel J, Hofreiter M, Foerster DWG (2020) Ancestral mitogenome capture of the Southeast Asian banded linsang. PLoS ONE 15(6): e0234385. https://doi.org/10.1371/journal.pone.0234385
Editor: Tzen-Yuh Chiang, National Cheng Kung University, TAIWAN
Received: July 31, 2019; Accepted: May 26, 2020; Published: June 30, 2020
Copyright: © 2020 Paijmans et al. 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.
Data Availability: Data is deposited in GenBank with the following Accession Numbers: MT559408-MT559415.
Funding: This work was supported by the European Research Council (consolidator grant GeneFlow #310763 to M.H.) and by the Leibniz-Association (grant SAW-2013-IZW-2 to JF). The NVIDIA TITAN-X GPU used for BEAST analyses was kindly donated by the NVIDIA Corporation.
Competing interests: The authors have declared that no competing interests exist.
Ancestral hybridisation capture is a method that employs RNA or DNA oligonucleotides (“baits”) based on a reconstructed ancestral sequence of the target region to selectively enrich genetic libraries prior to sequencing, thus increasing target sequence output compared to standard shotgun sequencing. As such, it can be employed in cases where no closely related reference sequence species is available, as has been shown by its successful application to recover a mitochondrial genome from the extinct glyptodont . Hybridisation capture in general is particularly suited to the genetic analysis of ancient and historical specimens, where DNA is likely to be highly degraded and potentially contaminated. Additional case studies on the application of ancestral capture are therefore useful to establish this approach as a standard method for the genetic investigation of species which lack (extant) close relatives.
Asiatic linsangs (Prionodon sp.) are an example of a genus for which little molecular data exists, and sampling fresh material is challenging. Linsangs are small, genet-like carnivores with a wide Southeast Asian distribution (Fig 1). The monogeneric family Prionodontidae consists of two extant species: the banded linsang (Prionodon linsang) that inhabits Sundaic southeast Asia, and the spotted linsang (Prionodon pardicolor) that inhabits much of the non-Sundaic southeast Asian mainland (Fig 1). While these species have been characterized morphologically , and some basic knowledge exists regarding their ecology [3–7], little is known about the populations of these elusive nocturnal carnivores. Molecular data has helped to resolve their phylogenetic relationship to other carnivorans—resulting in the establishment of their own family (Prionodontidae), rather than being placed within the Viverridae [8–10]. However, very limited investigation of intraspecific genetic variation has been done for the spotted linsang , and equivalent data for the banded linsang is lacking entirely. Furthermore, there is no knowledge on how populations are structured for either linsang species.
In order to assess the intraspecific variation of the banded linsang, we investigated specimens sampled at natural history museums (30–100 years old) from several parts of its Sundaic distribution (Sumatra, Borneo, and Java). We utilised a capture approach based on a reconstructed ancestral mitogenome (henceforth “ancestral capture”) sequence of the Felidae (the closest related family, ca. 28 million years (Ma) divergence from the Prionodon lineage; from Paijmans et al. ) to enrich the linsang genetic libraries for mitochondrial DNA sequences. The recovered sequences provided several novel insights into the evolution of the Asiatic linsangs. We also evaluated the efficiency of our capture strategy for regions of the mitogenome with high and with low sequence similarity, leading to recommendations for future ancestral capture experiments.
Materials & methods
Epithelial tissue from skulls or skins, or maxillo-turbinal bones, were obtained for ten historical banded linsang samples from natural history museums, eight of which yielded DNA (Table 1) and two failed to yield any mitochondrial DNA (S1 Table of S1 Data). The successful specimens originated from Borneo (N = 3), Sumatra (N = 2), Bangka Island (N = 2) and Java (N = 1). The provenance of one of the two Sumatran samples (PLI-5) has been drawn into question, as the collector of this specimen (M. Bartels) was primarily known for his work in Java, and only rarely visited Sumatra. Therefore, it is possible that this sample could represent an individual from Java instead of Sumatra.
DNA extraction & library preparation.
All pre-PCR steps were performed in dedicated ancient DNA facilities, with appropriate decontamination procedures in place. Digestion was performed using a non-destructive extraction buffer consisting of 5M of GuSCN, 25 nM NaCl, 50 mM of Tris, 20 mM of EDTA, 1% Tween-20 and 1% ß-mercaptoethanol , and binding and washing steps were performed following Dabney et al. . Single-stranded DNA libraries [15,16] and double-stranded DNA libraries [17,18] were prepared from the resulting extracts. For the majority of samples, endogenous content was estimated using low-level shotgun sequencing (appr. 1 million reads per library, 75bp paired- and single-end reads) prior to enrichment. For sample PLI-12, additional shotgun sequencing was performed in order to de-novo assemble a complete mitochondrial genome for the banded linsang (appr. 10 million reads; Table 1; S1 Table of S1 Data; also in Taron et al. ).
Ancestral mitogenome enrichment.
Sequence capture was performed using a 244k feature SureSelect microarray (Agilent), following the procedure described by Paijmans et al. . A reconstructed ancestral mitogenome of the carnivoran family Felidae  was used to design bait sequences using 2 bp tiling. Although a more closely related mitogenome sequence has recently become available (spotted linsang, P. pardicolor ), this sequence was not published at the time the current study was initiated, thus the ancestral sequence assay was utilised . Libraries were pooled prior to enrichment according to their target content as well as their molarity (S2 Table of S1 Data), to avoid libraries with high mitogenome content overwhelming samples with low mitogenome content on the array. After enrichment, approximately 30 million 75 bp paired-end reads were generated using the Illumina NextSeq 500 sequencing platform (S1 Table of S1 Data), using custom sequencing primers as needed for single-stranded libraries [15,21].
Raw sequences were trimmed using SeqPrep for paired-end sequence data (available from https://github.com/jstjohn/SeqPrep) and cutadapt v1.10  for single-ended sequence data, both with default parameters. Reads shorter than 30 bp after trimming were discarded.
Mitogenome assembly & mapping.
MITObim (Hahn et al. 2013) was used to assemble the P. linsang mitogenome, using the mitogenome from P. pardicolor (GenBank Acc. Nr. NC_024569 ) as reference sequence and a maximum mismatch value of 15%. After alignment, duplicates were identified according to both the 5’ and 3’ end mapping coordinates using MarkDuplicatesByStartEnd.jar (https://github.com/dariober/Java-cafe/tree/master/MarkDupsByStartEnd). To improve the accuracy of the resulting mitogenome further, we re-mapped the sequence data to the mitobim reference using the Burrows-Wheeler Aligner (BWA) v0.7.8 , with default values for seed length (32 bp) and mismatch values (0.04). Samtools v1.19  was used to remove reads with a mapping quality < Q30. Duplicates were again identified and removed with MarkDuplicatesByStartEnd.jar. The final consensus banded linsang mitogenome sequence was then called from the mapped data, using a minimum read depth of 3x and a 90% majority rule for base calling. By using this stringent base calling strategy, any positions with inadvertent enrichment of nuclear copies of mitochondrial origin (numts) should be detected and not incorporated into the final consensus. Annotation was performed manually in Geneious v7.0 , using the published mitogenome sequence of the spotted linsang P. pardicolor (GenBank Acc. Nr. NC_024569 ) as reference. For linsang samples not selected for de-novo assembly, shotgun and capture sequences were mapped to the new mitobim-assembled banded linsang mitogenome, using the same mapping tools and parameters as described above.
Sliding window analysis.
Sliding window sequence identity between the ancestral bait sequence and the P. linsang sequence was determined using a custom perl script, in 60 bp windows with a step-size of 30 bp. Due to insertion/deletion differences between the ancestral mitogenome and the linsang in the rRNA’s and the control region, this analysis excluded those regions and was restricted to the region between 2–15 kb. Sequencing depth in each window was recovered using bedtools v2.23 . To assess the impact of sequence divergence on capture efficiency, we compared the shotgun data from individual PLI-12 to the capture data from the same individual. For this we investigated the relative proportions of the number of 60 bp windows and sequencing depth (capture/shotgun) at different levels of sequence divergence from the ancestral bait sequence. These metrics were plotted using ggplot2  using the statistical programming environment R (http://www.cran.r-project.org).
Linsang sequences were aligned using ClustalW  as implemented in Geneious v7.0, with default parameters. For interspecific analyses, a second alignment was generated for the fossil calibration analysis (see below) following the same procedure, including the two most divergent banded linsang sequences (one from each matriline; PLI-5 and PLI-20) and 21 additional Feliformia species (S1 Data). All alignments were checked for internal stop codons using the vertebrate mitochondrial translation table in Geneious, to further ensure no numt sequences were integrated in the mitochondrial genome sequences. Furthermore, all columns with missing data were removed from the alignment.
- Analysis 1 involved maximum-likelihood analysis using RaXML v8.2.4  of representative mitochondrial sequences from the Feliformia clade, with the aim of reconstructing their phylogeny and determining the phylogenetic position of the linsang sequences.
- Analysis 2 involved analysis of the Feliformia dataset using BEAST v1.8.2 , with the aim of estimating the divergence time of banded and spotted linsangs, and the basal divergence of the sampled haplotype lineages within the banded linsang, using a total of eight fossil calibrations (S4 Table of S1 Data). The latter was achieved by the inclusion of two sequences representing the maximal phylogenetic divergence observed within the sampled specimens based on prior exploratory analyses.
- Analysis 3 involved analysis of all eight banded linsang sequences in BEAST, with the aim of estimating coalescence times of the sampled mitochondrial lineages, based on the tMRCA estimated in Analysis 2.
- Analysis 4 involved generating a median-joining network for the banded linsang samples using PopArt v1.7 .
High quality mitogenome assembly for the banded linsang
Approximately 8 million reads of shotgun data for sample PLI-12 were generated in order to assemble the mitogenome of the banded linsang with MITObim . We recovered 55,442 unique reads using mitobim after 24 iterations. Conventional mapping (BWA) of the shotgun data to the mitobim sequence resulted in 53,797 unique reads (Table 1; S1 Table of S1 Data), which were then used to generate the consensus mitogenome sequence. Using this approach, 99% of the mitogenome was reconstructed, with an average sequencing depth of 156x. We found that the banded linsang mitogenome had 88% sequence similarity to the published mitogenome sequence of the spotted linsang. The annotated banded linsang mitogenome sequence has been deposited in GenBank (accession number MT559410).
Ancestral capture efficiency at variable sequence similarity
In addition to the complete mitogenome, we furthermore recovered partial mitochondrial genomes for an additional 7 banded linsangs (Table 1): between 50–90% of these mitogenome were retrieved with high sequencing depth (39-275x; Table 1). Regions with little to no sequence divergence from baits showed proportionally higher read depth when compared to the shotgun data (Fig 2; S1 Fig of S1 Data). The uneven coverage along the mitogenome following ancestral capture thus likely reflects local sequence divergence between bait and target, a general property of hybridization capture that has been observed in multiple studies [e.g. 12,33–35]. The de-novo assembled banded linsang mitogenome showed between 0% and 27% sequence divergence to the ancestral bait sequence, with a mean of 12%.
Impact of sequence divergence between bait and target on coverage in 60 bp sliding windows (step size of 30 bp). The proportion of the number of windows per sequence divergence bin is given by the grey bars; trendlines (loess smooth regression) for the proportion of unique mapped reads per sequence divergence bin are shown separately for captured libraries (orange) and the shotgun library (black). The shotgun library shows no bias towards/against windows with high or low divergence, whereas the capture libraries show a very clear bias towards high identity windows.
Maximum likelihood analysis of mitochondrial sequences (23 sequences, 8001 bp) representing the Feliformia clade recovered the monophyly of the spotted linsang and the banded linsangs, and their position as sister to the Felidae clade, with 100% bootstrap support (S2 Fig of S1 Data). The median estimated age for the divergence of the two linsang species was 11.76 million years (Ma; 95% credibility interval [CI]: 9.44–14.40 Ma; Fig 3A). The median estimated age for the basal divergence of the banded linsang lineages was 1.22 Ma (95% CI: 0.85–1.60 Ma; Fig 3A). The intraspecific phylogeny (8 sequences, 8107 bp; Fig 3B; S2 Text of S1 Data) and MJ network (S3 Fig of S1 Data) reveal two divergent matrilines among the sampled banded linsang. Based on the estimated basal divergence of the banded linsang haplotype lineages, all other coalescence events among the eight sampled haplotypes are estimated to have occurred within the last ~500,000 years (Fig 3B).
A) Calibrated interspecific mitogenome phylogeny including a range of feliformia species. Calibrated nodes are indicated with green circles (S4 Table of S1 Data), and the ancestral bait sequence used for capture is indicated with a yellow star. Colours for the linsang taxa correspond to the distribution displayed in Fig 1. B) Calibrated mitogenome phylogeny of the banded linsang. On both panels, node labels indicate the Bayesian Posterior Probability, blue node bars indicate the 95% credibility interval of the posterior sampling of the node. Bottom axis indicates the age in millions of years.
Using a combination of shotgun sequencing and ancestral sequence capture, we were able to recover a complete mitogenome and seven additional partial mitogenomes of the banded linsang from historical museum specimens. This represents the first intraspecific mitochondrial dataset for this species, and provides novel insights into their evolution.
Phylogenetic analysis supports a deep divergence within the Asiatic linsangs, with the two recognised species coalescing around 12 Ma. This deep divergence between the spotted and banded linsang has been shown previously using short mtDNA sequences [36,37]. It is notable that this divergence approximates that found among all living felid species . Unfortunately, our small sample size and the questionable provenance of one of the Sumatran samples precludes any robust interpretation of the intraspecific population structure of the banded linsang. The potential presence of two divergent matrilines within the banded linsangs could serve as a basis for further, more thorough, investigation of the population structure using more samples and more genetic markers.
The problematic provenance of one of the Sumatran samples highlights both the advantages but also the difficulties of working with museum samples. As methodologies to mediate the less-than-favourable sample quality develop further and further, museum collections become an increasingly accessible and important sample source, providing access to genetic material of rare or extinct lineages that are otherwise not readily available. However, the provenance of museum samples is sometimes ambiguous, specifically regarding uncertainties about the geographic origin of the sample. Despite these issues, museum samples will remain a vital resource for rare species such as the Asiatic linsang.
Our study also provides further useful information on the practical application of ancestral capture. As the Felidae clade for which we reconstructed the ancestral mitogenome sequence is sister to the linsang lineage, we show that ancestral capture can be applied to enrich not only a range of descendant  as well as sister taxa (this study). Pooling of multiple samples in the same capture experiments will greatly leverage the sample throughput and cost-efficiency. However, in previous studies, pooling multiple libraries in the same capture experiment has sometimes led to uneven representation of each library in the post-capture pool [e.g. 39]. We find that pooling samples on the array according to their target content (based on low level shotgun sequencing) as well as their molarity, provided a relatively equal representation of reads per library in the post-capture pool (S1 Table of S1 Data). We therefore recommend this approach for pooling samples of variable sample quality in the same capture experiment in order to increase throughput and to decrease per-sample costs. While we carried out the capture on Agilent SureSelect arrays, this pooling strategy is likely to be beneficial to in-solution capture approaches as well.
Previous studies have found a strong correlation between enrichment efficiency and bait-to-target similarity [e.g. 12,33,34,40], while successful enrichment has been reported for up to 40% sequence divergence between bait and target [e.g. 41]. The maximum sequence divergence between our bait and target sequences was less than 30%. The large difference we observed between highly successful recovery at high-similarity regions (more than 1000x read depth in regions <10% sequence divergence) but low recovery at low-similarity regions (down to 0x depth in regions >20% sequence divergence) suggests a strong bias towards the former, in our case severe enough to lose the latter regions entirely. Some of this bias is expected due to the increased mismatches between bait and target; however, the lack of recovered data at higher levels of sequence divergence is not expected. It is possible that this may have resulted from a template abundance bias in post-capture amplification steps, and/or the two consecutive capture rounds. Although a single capture round has been shown to be less effective for overall enrichment rates compared with two consecutive capture rounds for degraded samples [42–44], the bias in recovery of low-similarity regions we observed may be partially reduced when only a single round of capture is used. An alternative solution could potentially be to split the capture baits into high- and low-identity pools, and capture the library separately with each bait pool—although not every bait preparation method will lend itself for this strategy (e.g. when preparing home-made baits from PCR products), and this also requires prior knowledge of local sequence divergence which may not be available.
Despite the inability of our ancestral capture strategy to recover more divergent mitochondrial regions in this experiment, overall we were able to recover a sufficient portion of the mitochondrial genome for phylogenetic and phylogeographic analyses. Thus, even without further optimisation, this approach is tenable for many applications including phylogenetic, phylogeographic and conservation genetic investigations of rare taxa. Furthermore, it is notable that this data can be generated rapidly and at relatively low cost, including for historical museum specimens, and is applicable even to the sister taxon of the ancestral bait sequence in addition to the descendent clade. Ancestral capture is thus a useful tool for genetic investigation of species which lack sequence information from close relatives.
We would like to thank the Naturalis Leiden (Netherlands), and the Sabah Museum and Sabah Parks (Malaysian Borneo) for contributing sample material. We thank Andreas Wilting for providing samples that he collected.
- 1. Delsuc F, Gibb GC, Kuch M, Billet G, Hautier L, Southon J, et al. The phylogenetic affinities of the extinct glyptodonts. Curr Biol. 2016;26: R155–R156. pmid:26906483
- 2. Hunter L. Carnivores of the world. Princeton University Press Princeton; 2011.
- 3. Lim BL. The banded linsang and the banded musang of West Malaysia. Malay Nat J. 1973;26: 105–111.
- 4. Van Rompaey H. The banded linsang, Prionodon linsang. Small Carniv Conserv. 1993;9: 11–15.
- 5. Kuznetzov GV, Baranauskas K. Notes on the behaviour, activity and feeding of the Spotted Linsang (Prionodon pardicolor). Small Carniv Conserv. 1993;8.
- 6. Gaubert P. Family Prionodontidae (linsangs). Handb Mamm World. 2009;1: 170–173.
- 7. Jennings AP, Veron G. Predicted distributions, niche comparisons, and conservation status of the spotted linsang (Prionodon pardicolor) and banded linsang (Prionodon linsang). Mammal Res. 2015;60: 107–116.
- 8. Philippe Gaubert, Géraldine Veron. Exhaustive sample set among Viverridae reveals the sister-group of felids: the linsangs as a case of extreme morphological convergence within Feliformia. Proc R Soc Lond B Biol Sci. 2003;270: 2523–2530. pmid:14667345
- 9. Gaubert P, Cordeiro-Estrela P. Phylogenetic systematics and tempo of evolution of the Viverrinae (Mammalia, Carnivora, Viverridae) within feliformians: implications for faunal exchanges between Asia and Africa. Mol Phylogenet Evol. 2006;41: 266–278. pmid:16837215
- 10. Hassanin A, Veron G. The complete mitochondrial genome of the Spotted Linsang, Prionodon pardicolor, the first representative from the family Prionodontidae (Mammalia, Carnivora). Mitochondrial DNA Part A. 2016;27: 912–913. pmid:24937573
- 11. Zhao J, Yao Y, Li D, Ni Q, Xie M, Zhang M, et al. The complete mitochondrial genome sequence of spotted linsang (Prionodon pardicolor pardicolor). Conserv Genet Resour. 2017;9: 177–180.
- 12. Paijmans JLA, Fickel J, Courtiol A, Hofreiter M, Förster DW. Impact of enrichment conditions on cross-species capture of fresh and degraded DNA. Mol Ecol Resour. 2016;16: 42–55. pmid:25925277
- 13. Rohland N, Siedel H, Hofreiter M. Nondestructive DNA extraction method for mitochondrial DNA analyses of museum specimens. Biotechniques. 2004;36: 814–820. pmid:15152601
- 14. Dabney J, Knapp M, Glocke I, Gansauge M-T, Weihmann A, Nickel B, et al. Complete mitochondrial genome sequence of a Middle Pleistocene cave bear reconstructed from ultrashort DNA fragments. Proc Natl Acad Sci. 2013;110: 15758–15763. pmid:24019490
- 15. Gansauge M-T, Meyer M. Single-stranded DNA library preparation for the sequencing of ancient or damaged DNA. Nat Protoc. 2013;8: 737–748. pmid:23493070
- 16. Korlević P, Gerber T, Gansauge M-T, Hajdinjak M, Nagel S, Aximu-Petri A, et al. Reducing microbial and human contamination in DNA extractions from ancient bones and teeth. BioTechniques. 2015;59. pmid:26260087
- 17. Meyer M, Kircher M. Illumina Sequencing Library Preparation for Highly Multiplexed Target Capture and Sequencing. Cold Spring Harb Protoc. 2010;2010: pmid:20516186
- 18. Henneberger K, Barlow A, Paijmans JLA. Double-Stranded Library Preparation for Ancient and Other Degraded Samples. In: Shapiro B, Barlow A, Heintzman PD, Hofreiter M, Paijmans JLA, Soares AER, editors. Ancient DNA. New York, NY: Springer New York; 2019. pp. 65–73. https://doi.org/10.1007/978-1-4939-9176-1_8 pmid:30875045
- 19. Taron UH, Lell M, Barlow A, Paijmans JLA. Testing of Alignment Parameters for Ancient Samples: Evaluating and Optimizing Mapping Parameters for Ancient Samples Using the TAPAS Tool. Genes. 2018;9. pmid:29533977
- 20. Paijmans JLA, González Fortes G, Förster DW. Application of Solid-State Capture for the Retrieval of Small-to-Medium Sized Target Loci from Ancient DNA. In: Shapiro B, Barlow A, Heintzman PD, Hofreiter M, Paijmans JLA, Soares AER, editors. Ancient DNA. New York, NY: Springer New York; 2019. pp. 129–139. https://doi.org/10.1007/978-1-4939-9176-1_14 pmid:30875051
- 21. Paijmans JLA, Baleka S, Henneberger K, Taron UH, Trinks A, Westbury MV, et al. Sequencing single-stranded libraries on the Illumina NextSeq 500 platform. ArXiv171111004 Q-Bio. 2017; Available: http://arxiv.org/abs/1711.11004
- 22. Martin M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.journal. 2011;17: 10–12.
- 23. Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009;25: 1754–1760. pmid:19451168
- 24. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics. 2009;25: 2078–2079. pmid:19505943
- 25. Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, et al. Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics. 2012;28: 1647–1649. pmid:22543367
- 26. Quinlan AR. BEDTools: The Swiss-Army Tool for Genome Feature Analysis. Curr Protoc Bioinforma. 2014;47: 11.12.1–11.12.34. pmid:25199790
- 27. Wickham H. ggplot2. Wiley Interdiscip Rev Comput Stat. 2011;3: 180–185.
- 28. Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, et al. Clustal W and Clustal X version 2.0. Bioinformatics. 2007;23: 2947–2948. pmid:17846036
- 29. Stamatakis A. RAxML Version 8: A tool for Phylogenetic Analysis and Post-Analysis of Large Phylogenies. Bioinformatics. 2014; btu033. pmid:24451623
- 30. Drummond AJ, Suchard MA, Xie D, Rambaut A. Bayesian Phylogenetics with BEAUti and the BEAST 1.7. Mol Biol Evol. 2012;29: 1969–1973. pmid:22367748
- 31. Leigh JW, Bryant D. popart: full-feature software for haplotype network construction. Methods Ecol Evol. 2015;6: 1110–1116.
- 32. Hahn C, Bachmann L, Chevreux B. Reconstructing mitochondrial genomes directly from genomic next-generation sequencing reads—a baiting and iterative mapping approach. Nucleic Acids Res. 2013;41: e129–e129. pmid:23661685
- 33. Portik DM, Smith LL, Bi K. An evaluation of transcriptome-based exon capture for frog phylogenomics across multiple scales of divergence (Class: Amphibia, Order: Anura). Mol Ecol Resour. 2016;16: 1069–1083. pmid:27241806
- 34. Peñalba JV, Smith LL, Tonione MA, Sass C, Hykin SM, Skipwith PL, et al. Sequence capture using PCR-generated probes: a cost-effective method of targeted high-throughput sequencing for nonmodel organisms. Mol Ecol Resour. 2014;14: 1000–1010. pmid:24618181
- 35. Richards SM, Hovhannisyan N, Gilliham M, Ingram J, Skadhauge B, Heiniger H, et al. Low-cost cross-taxon enrichment of mitochondrial DNA using in-house synthesised RNA probes. PLOS ONE. 2019;14: e0209499. pmid:30716066
- 36. Nyakatura K, Bininda-Emonds OR. Updating the evolutionary history of Carnivora (Mammalia): a new species-level supertree complete with divergence time estimates. BMC Biol. 2012;10: 12. pmid:22369503
- 37. Gaubert P, Tranier M, Delmas A-S, Colyn M, Veron G. First molecular evidence for reassessing phylogenetic affinities between genets (Genetta) and the enigmatic genet-like taxa Osbornictis, Poiana and Prionodon (Carnivora, Viverridae). Zool Scr. 2004;33: 117–129.
- 38. Johnson WE, Eizirik E, Pecon-Slattery J, Murphy WJ, Antunes A, Teeling E, et al. The Late Miocene Radiation of Modern Felidae: A Genetic Assessment. Science. 2006;311: 73–77. pmid:16400146
- 39. Fortes GG, Grandal-d’Anglade A, Kolbe B, Fernandes D, Meleg IN, García-Vázquez A, et al. Ancient DNA reveals differences in behaviour and sociality between brown bears and extinct cave bears. Mol Ecol. 2016;25: 4907–4918. pmid:27506329
- 40. Vallender EJ. Expanding whole exome resequencing into non-human primates. Genome Biol. 2011;12: R87. pmid:21917143
- 41. Li C, Hofreiter M, Straube N, Corrigan S, Naylor GJ. Capturing protein-coding genes across highly divergent species. BioTechniques. 2013;54: 321–326. pmid:23758502
- 42. Templeton JEL, Brotherton PM, Llamas B, Soubrier J, Haak W, Cooper A, et al. DNA capture and next-generation sequencing can recover whole mitochondrial genomes from highly degraded samples for human identification. Investig Genet. 2013;4: 1–13. pmid:23286546
- 43. White LC, Fontsere C, Lizano E, Hughes DA, Angedakin S, Arandjelovic M, et al. A roadmap for high-throughput sequencing studies of wild animal populations using noninvasive samples and hybridization capture. Mol Ecol Resour. 2019;19: 609–622. pmid:30637963
- 44. Förster DW, Bull JK, Lenz D, Autenrieth M, Paijmans JLA, Kraus RHS, et al. Targeted resequencing of coding DNA sequences for SNP discovery in nonmodel species. Mol Ecol Resour. 2018;18: 1356–1373. pmid:29978939
- 45. Duckworth, J.W., Mathai, J., Chutipong, W., Brodie, J. & Wilting, A. Prionodon linsang. The IUCN Red List of Threatened Species 2016: e.T41705A45219711. http://dx.doi.org/10.2305/IUCN.UK.2016-1.RLTS.T41705A45219711.en.
- 46. Duckworth, J.W., Lau, M., Choudhury, A., Chutipong, W., Timmins, R.J., Willcox, D.H.A., et al. 2016. Prionodon pardicolor. The IUCN Red List of Threatened Species 2016: e.T41706A45219917. http://dx.doi.org/10.2305/IUCN.UK.2016-1.RLTS.T41706A45219917.en.