Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Benchmark
Comparison and optimization of protocols and
whole-genome capture conditions for ancient
DNA samples
Reyhan Yaka*,1,2,3 ,MajaKrzewi
´
nska‡,1,2, Vendela Kempe Lagerholm‡,1,2, Anna Linderholm1,4,F
¨
usun ¨
Ozer5, Mehmet Somel3&
Anders G¨
otherstr¨
om1,2
1Centre for Palaeogenetics, Stockholm, Sweden; 2Department of Archaeology & Classical Studies, Stockholm University, Stockholm, Sweden; 3Department of Biological
Sciences, Middle East Technical University (METU), Ankara, Turkey; 4Department of Geological Sciences, Stockholm University, Stockholm, Sweden; 5Department of
Anthropology, Hacettepe University, Ankara, Turkey; *Author for correspondence: reyhan.yaka@su.se; ‡These authors contributed equally
BioTechniques 76: 00–00 (May 2024) 10.2144/btn-2023-0107
First draft submitted: 2 November 2023; Accepted for publication: 1 March 2024; Published online: 26 March 2024
ABSTRACT
Ancient DNA (aDNA) obtained from human remains is typically fragmented and present in relatively low amounts. Here we investigate a set of
optimal methods for producing aDNA data by comparing silica-based DNA extraction and aDNA library preparation protocols. We also test the
efficiency of whole-genome enrichment (WGC) on ancient human samples by modifying a number of parameter combinations. We find that the
Dabney extraction protocol performs significantly better than alternatives. We further observed a positive trend with the BEST library protocol
indicating lower clonality. Notably, our results suggest that WGC is effective at retrieving endogenous DNA, particularly from poorly-preserved
human samples, by increasing human endogenous proportions by 5x. Thus, aDNA studies will be most likely to benefit from our results.
METHOD SUMMARY
We studied a set of methods used to generate ancient genome data from archaeological human samples from different sites and periods, and
with varying levels of preservation. In these experiments we compared two sample preparation techniques, two silica-based DNA extraction
protocols, and two aDNA library preparation protocols. Finally, we explored the efficiency of whole-genome capture (WGC) on aDNA libraries
with different characteristics by comparing a set of protocol modifications.
KEYWORDS:
ancient DNA •DNA extraction •aDNA library •whole genome capture •optimization
Ancient DNA (aDNA) data obtained from archaeological remains an expanding and well-established field that has already helped resolve
various questions in evolutionary biology and anthropology. However, aDNA also exhibits significant technical challenges. For example,
authentic aDNA is retrieved as damaged and degraded into short fragments, and mixed with large amounts of bacteria and other en-
vironmental DNA [1–3]. These properties impact the efficiency of shotgun sequencing of ancient genomes. Although high-throughput
sequencing and in-solution capture techniques have helped to partly overcome these challenges, most archaeological samples still re-
main too badly preserved for in-depth genomic analysis. Over the last decade, a number of new approaches have mitigated these issues,
i.e., improvements in DNA extraction and library preparation [4–6]. But obtaining sufficient authentic aDNA still remains a challenge, and
this particularly holds true for samples from temperate, warm, humid areas, and acidic soils [7].
Hybridization-based capture approaches are explored to increase the data yield and decrease the cost. The most frequently used is
targeted single nucleotide polymorphism (SNP) capture; although promising, it relies on pre-ascertained SNP data. This poses disadvan-
tages over analyzing whole-genome shotgun data by limiting the set of statistical analyses applicable to genomic data and by suffering
from downstream biases. Therefore, a whole-genome capture (WGC)-based method could be a promising alternative [8,9]. Recently,
WGC has been used to generate genomic data from Neolithic and Bronze Age samples [10,11]. However, the popularity of WGC seems to
have declined lately, mainly because of high clonality levels observed in these libraries. To our knowledge, though, no systematic WGC
optimization trial has yet been performed on samples from various periods with various preservation levels.
Here, we aimed to develop an optimized set of methods to generate genomic data from poorly preserved human samples, building on
the original protocols. We studied three stages of aDNA data production: DNA extraction, library preparation and whole-genome enrich-
ment (Figure 1). We used archaeological samples from different sites and periods, especially from warm areas, and with varying levels of
preservation. We compared two sampling methods, drilling versus freezer-milling; two protocols for silica-based aDNA extraction, Ottoni
et al. [12]versus Dabney et al. [13]; and two library preparation protocols, BEST [14]versus M&K [15]. We also explored different conditions
used for a WGC approach and thus optimized DNA enrichment on ancient human samples. We used five sequencing summary statis-
Vol. 76 No. 5 C
2024 Stockholm University www.BioTechniques.com
1
Benchmark
Sample preparation
(pulverizing)
Drill 120 mg Freezer-mill 120 mg Freezer-mill 200 mg
DNA extraction Dabney Ottoni Dabney Ottoni Dabney Ottoni
aDNA library preparation BEST M&K
DNA extract
(Freezer-mill 120 mg and Dabney)
A
C
aDNA library
(Freezer-mill 120 mg,
Dabney and M&K)
Capture
Hybridization
temperature – time
Capture
Polymerase enzyme
60°C – 48 h 60°C – 66 h 65°C – 48 h 65°C – 66 h
Herculase KAPA Herculase KAPA Herculase KAPA Herculase KAPA
Capture
Input DNA concentration 100 ng/7 µl 200 ng/7 µl
aDNA library
(Dabney and M&K)
B
D
Figure 1. Experimental design. (A) For the comparisons of sample preparation and DNA extraction methods. This was applied to n =2 samples. (B) For
the comparisons of aDNA library protocols: BEST and M&K. This was applied to n =27 samples. (C) For the input DNA concentration tests used in
whole-genome capture. This was applied to n =3 samples. (D) For the comparisons and optimization of the whole-genome capture conditions. This
was applied to n =15 samples.
tics for all comparisons as evaluators: endogenous human DNA proportion (excluding duplicate reads), clonality, average read length,
GC-content, and the normalized number of SNPs covered at least once (Supplementary Appendix 1).
We first compared sample preparation (pulverizing) techniques and extraction protocols, exploring different combinations and using
n = 2 samples. We find that the Dabney [13]DNA extraction protocol performs significantly better compared to the Ottoni [12]protocol
with respect to human DNA proportion (median = 3.7×higher with Dabney), clonality and number of SNPs (Wilcoxon signed-rank test
[WSRT] p <0.05; Figures 2 & 3A & Supplementary Figure 1). We observed no significant difference between two protocols based on
average read length (WSRT p = 0.69; Supplementary Table 9). GC-content of sequenced reads was on average higher using Ottoni
(Dabney median = 38.7%; Ottoni median = 45.1%), which might be related to the higher clonality. Our results further suggest that freezer-
milling is a better technique than drilling for sample preparation (Figure 3B & Supplementary Figure 1), which is consistent with an earlier
study [16]. Although not significant, there is a positive trend toward higher yields with freezer-milling (WSRT p >0.1; Supplementary
Table 9). However, we also note that the observed difference in performance between freezer-milling and drilling reported in this study
could be minimized by using even lower drilling speeds, as reported in Adler et al. [16].
We next compared two aDNA library preparation protocols, M&K [15]and BEST [14], using n = 27 samples from various archaeolog-
ical periods (Supplementary Appendix 1). Our results showed nonsignificant differences between the BEST and M&K protocols when
all samples (n = 27) were considered with respect to human DNA proportion, average read length, GC-content and number of SNPs.
However, the BEST protocol resulted in significantly lower clonality (WSRT p <0.0001; BEST median = 18.2%; M&K median = 26.6%).
Also, BEST showed a positive trend with respect to both human DNA proportion and number of SNPs, although nonsignificant (Figure 4A
& Supplementary Figure 1 & Supplementary Table 9). Likewise, we observed the same pattern in within-period comparisons (e.g., poorly
preserved early Neolithic samples [n = 17]) for clonality, human DNA proportion and number of SNPs, as we did in all-sample compar-
Vol. 76 No. 5 C
2024 Stockholm University www.BioTechniques.com
2
Normalized SNP number
*
0
10
20
30
40
50
Clonality
*
0
10
20
30
Human proportion %
Dabney - Drill 120mg
Dabney - Freezer Mill 120mg
Dabney - Freezer Mill 200mg
Ottoni - Drill 120mg
Ottoni - Freezer Mill 120mg
Ottoni - Freezer Mill 200mg
0
20
40
60
80
AEN1 AEN2AEN1 AEN2 AEN1 AEN2
Average read length
*
0
10
20
30
40
GC content
0.000
0.001
0.002
0.003
*
AEN1 AEN2 AEN1 AEN2
A
D E
BC
Figure 2. Five summary statistics of the sequence data obtained by applying each combination of pulverization method and extraction protocol to two
samples: AEN1 and AEN2. (A) Human DNA proportion (%). (B) Clonality. (C) Average read length. (D) GC-content (%). (E) Number of single nucleotide
polymorphisms. The significant results are marked with one asterisk (p <0.05); 120 mg and 200 mg sample powder obtained using drill and
freezer-mill.
Dabney: Dabney et al.’s aDNA extraction protocol; Ottoni: Ottoni et al.’s aDNA extraction protocol.
isons (Figure 4B, Supplementary Figure 2 & Supplementary Table 9). Our interpretation is that BEST causes lower DNA loss, most likely
due to the single tube technique and fewer cleanup steps involved in the protocol. We also suggest that the lower clonality of BEST
libraries, when combined with the relatively higher human DNA content, allows for deeper sequencing.
We further investigated the performance of WGC on aDNA libraries using the myBaits Human Whole Genome Capture Kit (Arbor
Biosciences, MI, USA). We aimed to identify the parameter settings that provide the most desirable results for WGC on aDNA using
samples with varying levels of preservation. We applied a set of modifications to the WGC parameters given in the aforementioned kit
protocol (Supplementary Appendix 1 & Figure 1). To assess the performance of WGC, we compared the same five summary statistics of
the enriched libraries with the screening results from the same libraries (Supplementary Tables 7 & 8). Across all conditions and libraries
(n = 126 tests), we observed modest increases after the enrichment in both human DNA proportion and number of SNPs (0.3×–60.4×,
median = 2.8×, and 0.06×–48×, median = 1.5×, respectively; Figures 5–7 & Supplementary Figure 5).
We first tested the effect of input DNA concentration levels, 100 ng/7μl versus 200 ng/7μl, using a library set (n = 3) from different
regions and periods (Supplementary Appendix 1). We found a trend toward better performance with 200 ng/7μl, with a slight difference
from 100 ng/7μl (Figure 5). However, the sample size was not sufficient to show significance, which may turn out to be replicable with
larger sample sizes.
We next studied the effects of six parameter settings and their different combinations on the efficiency of aDNA WGC: two different
polymerases used at the amplification step of the enriched libraries: KAPA HiFi HotStart Polymerase (Kapa Biosystems) and Herculase
II Fusion DNA Polymerase (Agilent Technologies), two hybridization temperatures (60◦Cand65
◦C) and two hybridization times (48 h
and 66 h). The set of libraries (n = 15) was chosen according to diverse values for human DNA proportion statistics including various
Vol. 76 No. 5 C
2024 Stockholm University www.BioTechniques.com
3
Benchmark
Dabney Ottonii Drill 120 mg Freezer Mill 120 mg
*
*
*
0.001
0.1
10
0.001
0.1
10
Human
proportion %
Clonality Average
read length
Normalized
SNP number
GC
content
Human
proportion %
Clonality Average
read length
Normalized
SNP number
GC
content
*
A B
Figure 3. Boxplots show comparisons of the methods with respect to five summary statistics obtained from the sequence data. (A) For the two
extraction protocols. (B) For the pulverization techniques (drill vs freezer-mill) using 120 mg bone powder as starting material (supplementary data).
The significant results are marked with one asterisk (p <0.05). The y-axes are log scaled.
BEST Meyer and Kircher
Human
proportion %
Clonality Average
read length
Normalized
SNP number
GC
content
Human
proportion %
Clonality Average
read length
Normalized
SNP number
GC
content
Early NeolithicAll samples
10-2
*
102
1
10-4
10-6
10-2
102
1
10-4
10-6
*
A B
Figure 4. Boxplots show comparisons of the BEST and M&K protocols with respect to five summary statistics obtained from the sequence data. (A)
All-sample comparisons from varying periods (n =27). (B) Poorly preserved samples from the early Neolithic period (n =17). The significant results are
marked with one asterisk (p <0.05). The y-axes are log scaled.
Vol. 76 No. 5 C
2024 Stockholm University www.BioTechniques.com
4
Fold - increase
GC content
Human proportion
Clonality
Average read length
Normalized SNP number
Capture 100 ng/7ul
Capture 200 ng/7ul
GC content
Human proportion
Clonality
Average read length
Normalized SNP number
Human
proportion
Clonality Average
read
length
Normalized
SNP
number
GC
content
Fold - increase
Capture 100 ng/7ul Capture 200 ng/7ul
AN SM1 SM2
0
2
4
6
1
3
5
2
4
6
1
3
5
AB
Figure 5. Five summary statistics of the whole-genome enriched libraries using 100 ng/7μl and 200 ng/7μl input DNA concentrations. Fold-increase
shown on the y-axis indicates an increase in the capture compared with the shotgun sequencing data (precapture) for (A) all-sample level and (B) per
sample level.
AN: Anatolian Neolithic; SM: Scandinavian Medieval
combinations (e.g., low human proportion with high clonality; Supplementary Table 8). Libraries were also chosen torepresent DNA from
five archaeological periods, such that at least two samples from each period were included (Supplementary Appendix 1).
First, we compared two different polymerases at the amplification step after the enrichment. We found that Herculase was more
efficient than KAPA with respect to human DNA proportion, clonality, average read length and number of SNPs (WSRT p <0.01; Figure 6
& Supplementary Table 9).
Second, when comparing hybridization temperatures (60◦Cand65
◦C), we found that 60◦C yielded a significantly higher human DNA
proportion and number of SNPs relative to 65◦C(WSRTp<0.0001). However, the difference in clonality, average read length and GC-
content between the two hybridization temperatures was nonsignificant (WSRT p >0.1; Figure 6 & Supplementary Table 9).
Third, when comparing hybridization time (48 h and 66 h), we found that the libraries incubated 66 h yielded significantly lower clonality
compared with the libraries incubated 48 h (WSRT p <0.01). Although our results present a positive trend toward the 66-h hybridization
time, we found no significant difference with respect to human DNA proportion, average read length, GC-content and number of SNPs
(Figure 6 & Supplementary Table 9).
Finally, we compared the four parameter combinations of hybridization temperature (60◦Cand65
◦C) and time (48 h and 66 h). We
found that the combination of 60◦C/66 h generated better results relative to the alternatives. This combination produced significantly
lower clonality and higher number of SNPs than 60◦C/48 h (WSRT p <0.001). Here, we also observed the same pattern based on
human DNA proportion, but nonsignificantly (60◦C/66 h median = 0.02; 60◦C/48 h median = 0.014; WSRT p >0.5). Besides, 60◦C/66 h
yielded significantly better results than both 65◦C/66 h and 65◦C/48 h with respect to human DNA proportion and number of SNPs
(WSRT p <0.001 and p <0.01, respectively), while there was no significant difference concerning other summary statistics: clonality,
average read length and GC-content. Our results suggest that the combination of 60◦C/66 h results in a higher number of SNPs across
all combinations (WSRT p <0.05). Furthermore, we did not observe any significant difference between the 65◦C/66 h and 65◦C/48 h
combinations (Figure 6 & Supplementary Table 9).
Our results show that the WGC approach can increase the sequencing efficiency, leading to significant increase in both human DNA
proportion and number of SNPs after enrichment. We also note that WGC on aDNA could work even with lower input DNA concentration
(100 ng/7μl) and lower human DNA proportions (<1%). Regarding hybridization temperature and time, 60◦C/66 h was identified as an
optimal combination. This combination provides more opportunities for hybridization, suggesting the main constraint is not background
noise but human DNA annealing. Thus, future optimization efforts could investigate the effect of different combinations (e.g., longer
hybridization and lower temperatures) than those recommended.
We further find that a precapture human DNA proportion of 1–5% led to modest enrichment in both human DNA proportion and
number of SNPs (2.5×–5×and 1.4×–2.9×, respectively; Figure 7 & Supplementary Figures 4 & 5). However, human DNA proportion and
Vol. 76 No. 5 C
2024 Stockholm University www.BioTechniques.com
5
Benchmark
Herculase KAPA
Enzyme 60 C 65 C
Hybridization temperature 48 hours 66 hours
Hybridization time
60 C - 48 hours
60 C - 66 hours
65 C - 48 hours
65 C - 66 hours
Hybridization temperature - time
Human
proportion
Clonality Average
read
length
Normalized
SNP
number
GC
content
Human proportion Clonality Average read length Normalized SNP numberGC content
*
*
*
*
0.3
1
3
10
30
0.3
1
3
10
30
Human
proportion
Clonality Average
read
length
Normalized
SNP
number
GC
content
Human
proportion
Clonality Average
read
length
Normalized
SNP
number
GC
content
*
*
*
***
**
*
Fold - increaseFold - increase
*
0.3
1
3
10
30
0.3
1
3
10
30
A
B
Figure 6. Relative efficiency of the whole-genome-enriched libraries with respect to five summary statistics. (A) Using the two polymerase enzymes
(Herculase and KAPA), two hybridization temperatures (60◦Cand65
◦C) and two hybridization times (48 h and 66 h), respectively. Statistically
significant results are marked with one asterisk (p <0.05). (B) Using different combinations of hybridization temperature and time (60◦C/48 h,
60◦C/66 h, 65◦C/48 h and 65◦C/66 h). Statistically significant results (p <0.05) are marked with asterisks: one asterisk for the comparisons of
60◦C/66 h versus 60◦C/48 h, two asterisks for the comparisons of 60◦C/66 h versus 65◦C/48 h and 65◦C/48 h and three asterisks for the comparisons
of 60◦C/66 h versus all three other combinations. The y-axes are log scaled.
number of SNPs could be enriched between 4.4×–35×(median = 9.6×)and3×–18.4×(median = 5.8×), respectively, within reasonable
clonality levels (21.1–70.5%), for the libraries with even lower precapture human DNA proportion (0.01–1%; Figure 7 & Supplementary
Figure 5). This contrasts with earlier studies that reported that a precapture DNA content of 1–25% was needed to result in effective
enrichment [17,18]. We also find that overall WGC increases clonality, average read length and GC-content to a low extent, as observed
in previous studies [9,18–20].
Overall, our results suggest that WGC is an efficient and cost-effective method for retrieving endogenous human DNA in samples of
various preservation levels. Besides, this method, in contrast to targeted SNP capture, has the potential to generate unbiased data and,
thereby, could present a viable option in future aDNA research targeting low-quality libraries.
Vol. 76 No. 5 C
2024 Stockholm University www.BioTechniques.com
6
Fold - increase
1
3
10
30
EN1
(0.34 %)
EN2
(0.01 %)
EN3
(0.99 %)
EN4
(0.01 %)
EN5
(9.57 %)
EN6
(0.94 %)
N1
(0.59 %)
N2
(0.45 %)
N3
(0.8 %)
BA1
(0.03 %)
BA2
(0.69 %)
IA1
(0.18 %)
IA2
(0.17 %)
R1
(3.59 %)
R2
(0.02 %)
Capture
Precapture
Human proportion
Clonality
Average read length
GC content
Normalized SNP number
Figure 7. Five summary statistics of each sample before and after the enrichment concerning the best combination across all the parameter
combinations for each sample. The best result for each sample was chosen across n =8 combinations (two polymerase enzymes [Herculase vs KAPA],
two hybridization temperatures [60◦Cvs65
◦C] and two hybridization times [66 h vs 48 h]). Precapture human DNA proportion of each library is given in
parentheses. The y-axis is log scaled.
BA: Bronze Age; EN: Early Neolithic; IA: Iron Age; N: Neolithic; R: Roman Period.
Supplementary data
To view the supplementary data that accompany this paper please visit the journal website at: www.future-science.com/doi/
suppl/10.2144/btn-2023-0107
Author contributions
R Yaka, VK Lagerholm and M Krzewinska designed the study with contributions by F ¨
Ozer, M Somel and A G¨
otherstr¨
om; R Yaka and VK
Lagerholm performed laboratory work; R Yaka conducted data analysis; R Yaka, M Krzewinska, VK Lagerholm, A Linderholm, F ¨
Ozer, M
Somel and AG wrote the manuscript. All authors have read and agreed to the published version of the manuscript.
Acknowledgments
The authors thank all colleagues in G¨
otherstr¨
om’s research group at the Centre for Palaeogenetics (CPG, Stockholm University) for
helpful discussion and Arielle R Munters, Torsten G¨
unther and Mattias Jakobsson (Department of Organismal Biology, Human Evolution,
Uppsala University) for processing raw sequencing data using their in-house aDNA pipeline.
Financial disclosure
This study was supported by ERC (consolidator grant no. 772390 to M Somel), EMBO (short-term fellowship grant no. STF 7909 to R
Yaka) and TUBITAK of Turkey (grant no. 117Z229 to M Somel). The analyses were partially performed by resources in the storage and
computing projects (NAISS 2023/23-262 and NAISS 2023/22-504) provided by the Swedish National Infrastructure for Computing (SNIC)
at Uppsala Multidisciplinary Center for Advanced Computational Science (UPPMAX). The authors have no other relevant affiliations or
financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials
discussed in the manuscript apart from those disclosed.
Competing interests disclosure
The authors have no competing interests or relevant affiliations with any organization or entity with the subject matter or materials
discussed in the manuscript. This includes employment, consultancies, stock ownership or options and expert testimony.
Writing disclosure
No writing assistance was utilized in the production of this manuscript.
Vol. 76 No. 5 C
2024 Stockholm University www.BioTechniques.com
7
Benchmark
Open access
This work is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 Unported License. To view a copy of this license, visit
http://creativecommons.org/licenses/by-nc-nd/4.0/
References
Papers of special note have been highlighted as: •of interest; •• of considerable interest
1. P¨
a¨
aboS,PoinarH,SerreDet al. Genetic analyses from ancient DNA. Annu Rev Genet. 2004;38:645–679. doi:10.1146/annurev.genet.37.110801.143214
2. Gilbert MTP, Bandelt HJ, Hofreiter M, et al. Assessing ancient DNA studies. Trends Ecol Evol. 2005;20(10):541–544. doi: 10.1016/j.tree.2005.07.005
3. Knapp M, Lalueza-Fox C, Hofreiter M, et al. Re-inventing ancient human DNA. Investig Genet. 2015;6:4. doi:10.1186/s13323-015-0020-4
4. RohlandN,GlockeI,Aximu-PetriA,et al. Extraction of highly degraded DNA from ancient bones, teeth and sediments for high-throughput sequencing. Nat Protoc. 2018;13:2447–2461.
doi: 10.1007/978-1-4939-9176-1 4
5. Glocke I, Meyer M. Extending the spectrum of DNA sequences retrieved from ancient bones and teeth. Genome Res. 2017;27:1230–1237. doi: 10.1101/gr.219675.116
6. Gansauge MT, Meyer M. Single-stranded DNA library preparation for the sequencing of ancient or damaged DNA. Nat Protoc. 2013;8(4):737–748. doi: 10.1038/nprot.2013.038
7. Shapiro B, Hofreiter M. A paleogenomic perspective on evolution and gene function: new insights from ancient DNA. Science 2014;343(6169):1236573. doi: 10.1126/science.1236573
8. Carpenter ML, Buenrostro JD, Valdiosera C, et al. Pulling out the 1%: whole-genome capture for the targeted enrichment of ancient DNA sequencing libraries. Am J Hum Genet.
2013;93(5):852–864. doi: 10.1016/j.ajhg.2013.10.002
9. Avila-Arcos MC, Sandoval-Velasco M, Schroeder H, et al. Comparative performance of two whole-genome capture methodologies on ancient DNA Illumina libraries. Methods Ecol Evol.
2015;6:725–734. doi: 10.1111/2041-210X.12353
10. Allentoft M, Sikora M, Sj¨
ogren KG, et al. Population genomics of Bronze Age Eurasia. Nature 2015;522:167–172. doi: 10.1038/nature14507
11. Yaka R, Mapelli I, Kaptan D, et al. Variable kinship patterns in Neolithic Anatolia revealed by ancient genomes. Curr Biol. 2021;31:2455–2468. doi: 10.1016/j.cub.2021.03.050
12. Ottoni C, Rasteiro R, Willet R, et al. Comparing maternal genetic variation across two millennia reveals the demographic history of an ancient human population in southwest Turkey. R
Soc Open Sci. 2016;3:150250. doi: 10.1098/rsos.150250
13. Dabney J, Knapp M, Glocke I, 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. doi: 10.1073/pnas.1314445110
•• Describes a well-established method for DNA extraction from ancient samples.
14. Carøe C, Gopalakrishnan S, Vinner L, et al. Single-tube library preparation for degraded DNA. MethodsEcolEvol.2018;9:410–419. doi: 10.1111/2041-210X.12871
•• Describes an excellent ancient DNA library preparation method for ancient samples.
15. Meyer M, Kircher M. Illumina sequencing library preparation for highly multiplexed target capture and sequencing. Cold Spring Harb Protoc. 2010;6:t5448. doi:10.1101/pdb.prot5448
•Introduces the first and commonly used ancient DNA library preparation protocol for ancient samples.
16. Adler CJ, Haak W, Donlon D, et al. Survival and recovery of DNA from ancient teeth and bones. J Archaeol Sci. 2011;38:956–964. doi: 10.1016/j.jas.2010.11.010
17. Avila-Arcos MC, Cappellini E, Romero-Navarro JA, et al. Application and comparison of large-scale solution-based DNA capture enrichment methods on ancient DNA. Sci Rep. 2011;1:74.
doi: 10.1038/srep00074
•Provides valuable information regarding whole-genome DNA capture methods.
18. Cruz-D´
avalos DI, Llamas B, Gaunitz C, et al. Experimental conditions improving in-solution target enrichment for ancient DNA. Mol Ecol Resour. 2016;17:508–522. doi: 10.1111/1755-
0998.12595
19. Ozga AT, Nieves-Col´
on MA, Honap TP, et al. Successful enrichment and recovery of whole mitochondrial genomes from ancient human dental calculus. Am.J.Phys.Anthropol.
2016;160:220–228. doi:10.1002/ajpa.22960
20. Ziesemer KA, Ramos-Madrigal J, Mann AE, et al. The efficacy of whole human genome capture on ancient dental calculus and dentin. Am.J.Phys.Anthropol.2019;168:496–509. doi:
10.1002/ajpa.23763
•Excellent comparative study regarding whole-genome capture method on ancient DNA.
Vol. 76 No. 5 C
2024 Stockholm University www.BioTechniques.com
8