The DOI matches its Crossref record; 28 of the 28 references it cites were found in a registry or an index; 12 of the 14 sources its claims cite were checked for support.
Unravelling reference bias in ancient DNA datasets
Open accessAbstract
Abstract Motivation The alignment of sequencing reads is a critical step in the characterization of ancient genomes. However, reference bias and spurious mappings pose a significant challenge, particularly as cutting-edge wet lab methods generate datasets that push the boundaries of alignment tools. Reference bias occurs when reference alleles are favoured over alternative alleles during mapping, whereas spurious mappings stem from either contamination or when endogenous reads fail to align to their correct position. Previous work has shown that these phenomena are correlated with read length but a more thorough investigation of reference bias and spurious mappings for ancient DNA has been lacking. Here, we use a range of empirical and simulated palaeogenomic datasets to investigate the impacts of mapping tools, quality thresholds, and reference genome on mismatch rates across read lengths. Results For these analyses, we introduce AMBER, a new bioinformatics tool for assessing the quality of ancient DNA mapping directly from BAM-files and informing on reference bias, read length cut-offs and reference selection. AMBER rapidly and simultaneously computes the sequence read mapping bias in the form of the mismatch rates per read length, cytosine deamination profiles at both CpG and non-CpG sites, fragment length distributions, and genomic breadth and depth of coverage. Using AMBER, we find that mapping algorithms and quality threshold choices dictate reference bias and rates of spurious alignment at different read lengths in a predictable manner, suggesting that optimized mapping parameters for each read length will be a key step in alleviating reference bias and spurious mappings. Availability and implementation AMBER is available for noncommercial use on GitHub (https://github.com/tvandervalk/AMBER.git). Scripts used to generate and analyse simulated datasets are available on Github (https://github.com/sdolenz/refbias_scripts).
Overview
CoChat Summary
## TL;DR
This paper introduces AMBER, a Python tool for ancient DNA quality control that reports mapping bias, damage, fragment length, and coverage statistics directly from BAM files. Using empirical and simulated datasets, the authors show that reference bias and spurious mappings depend strongly on aligner choice, map-quality thresholds, read length, and sample-reference divergence.
## Key findings
- AMBER computes four main ancient-DNA statistics at once: mismatch rates by read length, cytosine deamination profiles at CpG and non-CpG sites, fragment length distributions, and genome breadth/depth of coverage.
- Across empirical and simulated data, mapping algorithms and MQ thresholds changed reference bias and spurious alignment in predictable ways; the authors conclude that optimized, read-length-specific mapping parameters are important to reduce these biases.
- In a mapping comparison using American mastodon reads mapped to the Asian elephant genome, Bowtie2 yielded the greatest proportion of mapped data, but at 30–40 bp BWA-mem showed a “precipitous increase” in reference bias and Bowtie2 recovered only two-thirds of the expected mismatches per bp; for reads of ~40–80 bp, Bowtie2-mapped reads had fewer mismatches than BWA-aln or BWA-mem, implying underestimation of divergence.
- With MQ filtering, a strict MQ ≥25 produced strong reference bias, most pronounced in the shortest reads; using MQ ≥1, reads below 65 bp were ~50% more divergent from the reference than when using MQ ≥30, and below ~32 bp, MQ ≥30 retained only reads without mismatches despite known sample-reference divergence.
## Methods
- The study presents AMBER, a Python3 command-line tool that takes up to six BAM files and computes read-mapping bias, damage, fragment length, and coverage summaries in a four-panel plot.
- Evaluation used seven empirical ancient-DNA datasets (including six paleogenomes and one ancient metagenome) spanning species, ages of 2 ka–2 Ma, endogenous content of 3%–70%, and coverage of 0.3–42.4×, plus simulated datasets from Asian elephant and Black rhinoceros genomes.
- Empirical data were mapped with BWA-aln, Bowtie2, or BWA-mem against species-specific or interspecific references, then filtered with MQ thresholds of 1, 20, 25, or 30; simulated reads were generated with Gargammel, with 18.1 million reads per reference genome and divergence levels from 0% to 15%.
## Limitations
- The paper states that AMBER should be used only as a first data-authentication check, because non-endogenous data can mimic expected ancient-DNA patterns and look authentic.
- The simulation design did not model an ancient-DNA damage profile or heterogeneity in evolutionary rate across the genome; this is explicitly stated in the Methods.
- Evident from the text: the empirical benchmarking is limited to seven datasets and a restricted set of aligners/MQ settings, and the authors note that some findings are tool- and sample-specific rather than universal.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures




Verification
Claims the paper makes
8 Valid · 1 Not valid · 1 Not checked · 41 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“However, reference bias and spurious mappings pose a significant challenge, particularly as cutting-edge wet lab methods generate datasets that push the boundaries of alignment tools.”
§ Unravelling reference bias in ancient DNA datasets > Abstract > Motivation
What the sources say
- Partly supportedcites Günther and Nettelblad 2019 sourceExperimental — accuracy not yet measured
The Conclusion says reference bias is pervasive in ancient DNA data and can create spurious population-genomic results. The Discussion also says some library-preparation techniques target shorter fragments, which affect mapping. The paper does not establish that cutting-edge wet-lab methods generate datasets that push alignment tools to their limits, and its finding of spurious results is not the same as a finding of spurious mappings. (LLM, conf=0.82)
- Partly supportedcites Orlando et al. 2021 sourceExperimental — accuracy not yet measured
The Results section says mapping to a single reference genome can introduce substantial reference bias, and its read-processing discussion says mapper performance varies with sample-preparation methods. It discusses mapping specificity, but does not specifically identify spurious mappings as a significant challenge or say that cutting-edge wet-lab methods generate datasets that push alignment tools to their limits. No page numbers are available in our copy. (LLM, conf=0.84)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“An excess of these substitutions arise near read termini due to the accumulation of post-mortem cytosine deamination (Briggs et al. 2007).”
§ Unravelling reference bias in ancient DNA datasets > 1 Introduction > 1.1 Tool description AMBER is specifically designed for the quality assessment of aDNA sequence data directly from BAM-files, allows up to six samples to be analysed together, and eliminates the need for additional file formats, datasets, or preprocessing steps. This user-friendly software provides comprehensive insights into the quality of aDNA data by offering several key functionalities. Firstly, AMBER computes base mismatches between reads and the reference, normalized by the read length. This allows the user to identify biases introduced during read trimming, merging, mapping and quality filtering, and to determine appropriate sample-specific read length cutoffs in order to mitigate such biases (van der Valk et al. 2021). Secondly, AMBER assesses aDNA damage patterns by calculating the rate of C-to-T substitutions across sequence reads. An excess of these substitutions arise near read termini due to the accumulation of post-mortem cytosine deamination (Briggs et al. 2007). By characterizing damage patterns, users can distinguish authentic aDNA from modern contaminants, thereby ensuring the integrity of the dataset (Jónsson et al. 2013, Skoglund et al. 2014). AMBER also includes measures of DNA damage specifically at CpG sites, thereby allowing for the authentication of aDNA libraries that are chemically treated to eliminate cytosine deamination damage patterns outside of CpG sites (USER-treatment) (Briggs et al. 2010). Thirdly, AMBER outputs the aDNA fragment length distribution, a critical aspect for assessing the DNA degradation profile. The fragment length distribution helps users to evaluate the extent of DNA fragmentation, estimate the average size of endogenous DNA molecules, and assess the suitability of the data for further downstream analyses. Finally, AMBER provides an estimation of the genome coverage, by measuring both the fraction of the target genome covered by sequence reads (breadth of coverage) and the average sequence depth across the genome. These metrics aid the user in determining the representativeness and completeness of their genomic data, guiding downstream analyses such as variant calling, genomic sex determination, or the authentication of the presence of species of interest in metagenomic data (Supplementary Texts S1 and S2; Supplementary Figs S1 and S2). The running time of AMBER is ∼5–100× faster than other currently available software (Supplementary Text S3; Supplementary Table S1). To demonstrate the utility of AMBER, we applied this new tool to multiple ancient genomes from diverse samples of different quality and age. Through these case studies, we showcase the capabilities of AMBER in assessing the quality of aDNA data and provide insights into different biases that can be introduced during the bioinformatic processing of ancient genomes.
What the sources say
- Supportedcites Briggs et al. 2007 source
The Results and Discussion report a pronounced excess of C-to-T substitutions at the 5′ ends of Neandertal DNA reads and attribute it to deaminated cytosines in the ancient template strands. The paper further concludes that cytosine deamination is more prevalent at molecule ends than in their interiors. (LLM, conf=0.92)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Without employing proper data quality control checks, downstream analyses and the reliability of the results obtained from ancient genomes can become significantly biassed (Günther and Nettelblad 2019, Orlando et al. 2021).”
§ Unravelling reference bias in ancient DNA datasets > 1 Introduction
What the sources say
- Supportedcites Günther and Nettelblad 2019 source
The Conclusion says reference bias can create spurious results in ancient-DNA population genomic analyses and stresses assessing and mitigating systematic biases. The Discussion identifies filtering, standardized processing, and controls as ways to avoid their effects. Together, these support the claim that inadequate quality checks can compromise downstream results. (LLM, conf=0.90)
- Supportedcites Orlando et al. 2021 source
The Post-mortem DNA damage removal section says damage-related misincorporations can bias sequence analyses and lead to incorrect conclusions. The Error rates section also explains that artefactual variants can affect population-level analyses and describes checks that help ensure valid downstream conclusions. (LLM, conf=0.92)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“By characterizing damage patterns, users can distinguish authentic aDNA from modern contaminants, thereby ensuring the integrity of the dataset (Jónsson et al. 2013, Skoglund et al. 2014).”
§ Unravelling reference bias in ancient DNA datasets > 1 Introduction > 1.1 Tool description AMBER is specifically designed for the quality assessment of aDNA sequence data directly from BAM-files, allows up to six samples to be analysed together, and eliminates the need for additional file formats, datasets, or preprocessing steps. This user-friendly software provides comprehensive insights into the quality of aDNA data by offering several key functionalities. Firstly, AMBER computes base mismatches between reads and the reference, normalized by the read length. This allows the user to identify biases introduced during read trimming, merging, mapping and quality filtering, and to determine appropriate sample-specific read length cutoffs in order to mitigate such biases (van der Valk et al. 2021). Secondly, AMBER assesses aDNA damage patterns by calculating the rate of C-to-T substitutions across sequence reads. An excess of these substitutions arise near read termini due to the accumulation of post-mortem cytosine deamination (Briggs et al. 2007). By characterizing damage patterns, users can distinguish authentic aDNA from modern contaminants, thereby ensuring the integrity of the dataset (Jónsson et al. 2013, Skoglund et al. 2014). AMBER also includes measures of DNA damage specifically at CpG sites, thereby allowing for the authentication of aDNA libraries that are chemically treated to eliminate cytosine deamination damage patterns outside of CpG sites (USER-treatment) (Briggs et al. 2010). Thirdly, AMBER outputs the aDNA fragment length distribution, a critical aspect for assessing the DNA degradation profile. The fragment length distribution helps users to evaluate the extent of DNA fragmentation, estimate the average size of endogenous DNA molecules, and assess the suitability of the data for further downstream analyses. Finally, AMBER provides an estimation of the genome coverage, by measuring both the fraction of the target genome covered by sequence reads (breadth of coverage) and the average sequence depth across the genome. These metrics aid the user in determining the representativeness and completeness of their genomic data, guiding downstream analyses such as variant calling, genomic sex determination, or the authentication of the presence of species of interest in metagenomic data (Supplementary Texts S1 and S2; Supplementary Figs S1 and S2). The running time of AMBER is ∼5–100× faster than other currently available software (Supplementary Text S3; Supplementary Table S1). To demonstrate the utility of AMBER, we applied this new tool to multiple ancient genomes from diverse samples of different quality and age. Through these case studies, we showcase the capabilities of AMBER in assessing the quality of aDNA data and provide insights into different biases that can be introduced during the bioinformatic processing of ancient genomes.
What the sources say
- Partly supportedcites Jónsson et al. 2013 sourceExperimental — accuracy not yet measured
The Abstract says damage-related misincorporation patterns are instrumental in authenticating ancient sequences versus modern contaminants, and the Introduction says mapDamage identifies these patterns. The Conclusion describes reducing errors in downstream analyses, but the paper does not establish that characterizing damage patterns ensures the integrity of a dataset. (LLM, conf=0.84)
- Partly supportedcites Skoglund et al. 2014 sourceExperimental — accuracy not yet measured
The Results (Model Outline) describe using damage patterns to assess whether a DNA fragment is ancient, and the Discussion reports reducing contamination to negligible levels. But the paper treats this as a likelihood-based distinction, not a guarantee of dataset integrity; the Discussion also recommends further authentication of retained reads. (LLM, conf=0.84)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“However, between ∼40 and 80 bp, which includes the majority of empirical data, we observe that reads mapped with Bowtie2 have fewer mismatches to the reference than those mapped with BWA-aln and BWA-mem (Fig. 2a and b, top left).”
§ Unravelling reference bias in ancient DNA datasets > 3 Results > 3.3 Assessing mapping strategies A wide range of mapping software and aDNA-specific mapping parameters to optimize sequence yield have been explored (Schubert et al. 2012, Martiniano et al. 2020, Poullet and Orlando 2020, Oliva et al. 2021). The optimal mapping parameters are often sample-specific, and dependent on the reference genome used, the type of samples being analysed, and the questions under investigation. A commonly used statistic for assessing optimal mapping strategies is the mapped read count, which disregards the effects of spurious mappings and reference bias. AMBER allows the evaluation of mapping strategies and how they affect the underlying data, helping the user to implement the optimal strategy for their specific case. We used AMBER to compare USER-treated sequencing reads of an American mastodon that was mapped to the Asian elephant genome using BWA-aln, BWA-mem and Bowtie2, and filtered with MQ ≥1. In this particular example, a ∼10 bp periodicity is also observed in the fragment length distribution, consistent with DNA fragmentation in the presence of histone-DNA complexes (Pedersen et al. 2014). To further explore the three alignment tools, we additionally used simulated elephant reads with 2% sequence divergence, which is comparable to the observed divergence between American mastodon and Asian elephant (Fig. 2a, top left). In both the empirical and simulated data, Bowtie2 resulted in the greatest proportion of mapped data (Fig. 2a and b). At short read lengths (30–40 bp), BWA-mem displays a precipitous increase in reference bias, Bowtie2 recovers only two-thirds of the expected mismatches per bp, whereas BWA-aln shows performance comparable to other read lengths (Fig. 2a and b, top left). However, between ∼40 and 80 bp, which includes the majority of empirical data, we observe that reads mapped with Bowtie2 have fewer mismatches to the reference than those mapped with BWA-aln and BWA-mem (Fig. 2a and b, top left). This implies that the divergence between American mastodon and Asian elephant would be significantly underestimated when analysing a Bowtie2 mapped genome. This reference bias is mirrored in the aDNA damage plot, whereby Bowtie2-mapped data appears to exhibit less damage than the BWA-mapped data (Fig. 2a, top right). In contrast, a step-down pattern, or pull to the reference, is observed at longer read lengths (>120 bp) with BWA-aln although this appears to not affect those mapped with BWA-mem (Fig. 2b, right). The read mismapping rate of simulated data is lower for both BWA-aln and BWA-mem as compared to Bowtie2 (Fig. 2c). However, the Bowtie2 mismapping rate is greatly reduced at MQ ≥20 (Fig. 2c, Supplementary Data S4).
plainstates an associationSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
23 more claims
Create a free account to see all 51 claims and every source verdict.
- It may state two separate claims as one. 18
- Probably not a substantial claim. 10
Citations
Cites
The Sequence Alignment/Map format and SAMtools
Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., Marth, G., Abecasis, G., Durbin, R., & 1000 Genome Project Data Processing Subgroup (2009) · Bioinformatics · doi:10.1093/bioinformatics/btp352 · Cited by 62,041
Fast and accurate short read alignment with Burrows-Wheeler transform
Li, H., & Durbin, R. (2009) · Bioinformatics · doi:10.1093/bioinformatics/btp324 · Cited by 58,607
Fast gapped-read alignment with Bowtie 2
Langmead, B., & Salzberg, S. L. (2012) · Nature Methods · doi:10.1038/nmeth.1923 · Cited by 56,400 · cited 1×
“3.5.1 (Langmead and Salzberg 2012) on the --sensitive setting in the default end-to-end mode as recommended for ancient samples (Poullet and Orlando2020), or BWA-mem v0.”
InfluentialMatplotlib: a 2D graphics environment
Hunter, J. D. (2007) · Computing in Science & Engineering · doi:10.1109/mcse.2007.55 · Cited by 39,249 · cited 1×
“AMBER runs entirely in python3, with the only dependencies being pysam (Heger et al. 2014) and matplotlib (Hunter 2007).”
Backgroundfastp: an ultra-fast all-in-one FASTQ preprocessor
Chen, S., Zhou, Y., Chen, Y., & Gu, J. (2018) · Bioinformatics · doi:10.1093/bioinformatics/bty560 · Cited by 29,721 · cited 1×
“23.2, and removed reads below a length of 20 bp (or 30 bp in the cases of American mastodon and Siberian unicorn) and those that could not be merged (Chen et al. 2018).”
BackgroundInfluentialA draft sequence of the Neandertal genome
Green, R. E., Krause, J., Briggs, A. W., Maricic, T., Stenzel, U., Kircher, M., Patterson, N., Li, H., Zhai, W., Fritz, M. H.-Y., et al. (2010) · Science · doi:10.1126/science.1188021 · Cited by 4,318 · cited 4×
“7.17 (Li and Durbin 2009) with ancient DNA-specific parameters (-l 16500 -n 0.01 -o 2) (Green et al. 2010), Bowtie2 v2.”
BackgroundInfluentialmapDamage2.0: fast approximate Bayesian estimates of ancient DNA damage parameters
Jónsson, H., Ginolhac, A., Schubert, M., Johnson, P. L. F., & Orlando, L. (2013) · Bioinformatics · doi:10.1093/bioinformatics/btt193 · Cited by 1,669 · cited 2×
“By characterizing damage patterns, users can distinguish authentic aDNA from modern contaminants, thereby ensuring the integrity of the dataset (J � onsson et al. 2013, Skoglund et al. 2014).”
BackgroundSelf-citationPatterns of damage in genomic DNA sequences from a neandertal
Briggs, A. W., Stenzel, U., Johnson, P. L. F., Green, R. E., Kelso, J., Prüfer, K., Meyer, M., Krause, J., Ronan, M. T., Lachmann, M., et al. (2007) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.0704665104 · Cited by 981 · cited 1×
“An excess of these substitutions arise near read termini due to the accumulation of post-mortem cytosine deamination (Briggs et al. 2007).”
BackgroundSeparating endogenous ancient DNA from modern day contamination in a Siberian Neandertal
Skoglund, P., Northoff, B. H., Shunkov, M. V., Derevianko, A. P., Pääbo, S., Krause, J., & Jakobsson, M. (2014) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1318934111 · Cited by 499 · cited 3×
“By characterizing damage patterns, users can distinguish authentic aDNA from modern contaminants, thereby ensuring the integrity of the dataset (J � onsson et al. 2013, Skoglund et al. 2014).”
BackgroundMethodRemoval of deaminated cytosines and detection of in vivo methylation in ancient DNA
Briggs, A. W., Stenzel, U., Meyer, M., Krause, J., Kircher, M., & Pääbo, S. (2010) · Nucleic Acids Research · doi:10.1093/nar/gkp1163 · Cited by 491 · cited 1×
“AMBER also includes measures of DNA damage specifically at CpG sites, thereby allowing for the authentication of aDNA libraries that are chemically treated to eliminate cytosine deamination damage patterns outside of CpG sites (USER-treatment) (Briggs et al. 2010).”
BackgroundTemporal patterns of nucleotide misincorporations and DNA fragmentation in ancient DNA
Sawyer, S., Krause, J., Guschanski, K., Savolainen, V., & Pääbo, S. (2012) · PLoS ONE · doi:10.1371/journal.pone.0034131 · Cited by 489 · cited 1×
“Ancient DNA typically consists of short fragments in the range of 10–150 bp (Sawyer et al. 2012).”
BackgroundOrlando, L., Allaby, R., Skoglund, P., Der Sarkissian, C., Stockhammer, P. W., Ávila-Arcos, M. C., Fu, Q., Krause, J., Willerslev, E., Stone, A. C., et al. (2021) · Nature Reviews Methods Primers · doi:10.1038/s43586-020-00011-0 · Cited by 411 · cited 3×
“These phenomena can occur when mapping to either a conspecific or closely related genome in which the alignment tool favours alleles found within the reference over alternative alleles (G € unther and Nettelblad 2019, Orlando et al. 2021).”
BackgroundSelf-citationMillion-year-old DNA sheds light on the genomic history of mammoths
van der Valk, T., Pečnerová, P., Díez-del-Molino, D., Bergström, A., Oppenheimer, J., Hartmann, S., Xenikoudakis, G., Thomas, J. A., Dehasque, M., Sağlıcan, E., et al. (2021) · Nature · doi:10.1038/s41586-021-03224-9 · Cited by 345 · cited 7×
“To avoid false inferences, aDNA studies often adopt a conservative minimal read length filter of 35 bp, potentially excluding a substantial portion of usable data (de Filippo et al. 2018, van der Valk et al. 2021).”
BackgroundMethodInfluentialSelf-citationA 2-million-year-old ecosystem in Greenland uncovered by environmental DNA
Kjær, K. H., Winther Pedersen, M., De Sanctis, B., De Cahsan, B., Korneliussen, T. S., Michelsen, C. S., Sand, K. K., Jelavić, S., Ruter, A. H., Schmidt, A. M. A., et al. (2022) · Nature · doi:10.1038/s41586-022-05453-y · Cited by 341 · cited 2×
“Further, as new boundaries for aDNA recovery beyond the million year time range are tested (van der Valk et al. 2021, Kjær et al. 2022, Fernandez-Guerra et al. 2023), the ability to robustly study increasingly damaged samples from diverged populations requires an assessment of current alignment…”
BackgroundImproving ancient DNA read mapping against modern reference genomes
Schubert, M., Ginolhac, A., Lindgreen, S., Thompson, J. F., AL-Rasheid, K. A., Willerslev, E., Krogh, A., & Orlando, L. (2012) · BMC Genomics · doi:10.1186/1471-2164-13-178 · Cited by 325 · cited 1×
“A wide range of mapping software and aDNA-specific mapping parameters to optimize sequence yield have been explored (Schubert et al. 2012, Martiniano et al. 2020, Poullet and Orlando 2020, Oliva et al. 2021).”
BackgroundSelf-citationGünther, T., & Nettelblad, C. (2019) · PLOS Genetics · doi:10.1371/journal.pgen.1008302 · Cited by 228 · cited 3×
“These phenomena can occur when mapping to either a conspecific or closely related genome in which the alignment tool favours alleles found within the reference over alternative alleles (G € unther and Nettelblad 2019, Orlando et al. 2021).”
BackgroundBenchmarking short sequence mapping tools
Hatem, A., Bozdağ, D., Toland, A. E., & Çatalyürek, Ü. V. (2013) · BMC Bioinformatics · doi:10.1186/1471-2105-14-184 · Cited by 223 · cited 1×
“The higher rate of mismapping seen with Bowtie2 compared to BWA has previously been noted (Hatem et al. 2013), including at higher MQ filter thresholds of up to 30 (Poullet and Orlando 2020).”
BackgroundGenome-wide nucleosome map and cytosine methylation levels of an ancient human genome
Pedersen, J. S., Valen, E., Velazquez, A. M. V., Parker, B. J., Rasmussen, M., Lindgreen, S., Lilje, B., Tobin, D. J., Kelly, T. K., Vang, S., et al. (2014) · Genome Research · doi:10.1101/gr.163592.113 · Cited by 170 · cited 1×
“In this particular example, a � 10 bp periodicity is also observed in the fragment length distribution, consistent with DNA fragmentation in the presence of histone-DNA complexes (Pedersen et al. 2014).”
BackgroundSelf-citationUnearthing Neanderthal population history using nuclear and mitochondrial DNA from cave sediments
Vernot, B., Zavala, E. I., Gómez-Olivencia, A., Jacobs, Z., Slon, V., Mafessoni, F., Romagné, F., Pearson, A., Petr, M., Sala, N., et al. (2021) · Science · doi:10.1126/science.abf1667 · Cited by 169 · cited 1×
“…focus on mapping reads using a third-allele reference, which allows both alleles at heterozygous sites to be considered equally when mapping, therefore greatly reducing the impact of reference bias, albeit at the cost of decreased mapping rates (G € unther and Nettelblad 2019, Vernot et al. 2021).”
Backgroundgargammel: a sequence simulator for ancient DNA
Renaud, G., Hanghøj, K., Willerslev, E., & Orlando, L. (2017) · Bioinformatics · doi:10.1093/bioinformatics/btw670 · Cited by 150 · cited 2×
“The proportion of mismapped reads was calculated by searching for reads whose mapping coordinates did not match their known coordinates of origin that were retained in the read header by Gargammel.”
MethodSelf-citationMartiniano, R., Garrison, E., Jones, E. R., Manica, A., & Durbin, R. (2020) · Genome Biology · doi:10.1186/s13059-020-02160-7 · Cited by 128 · cited 1×
“A wide range of mapping software and aDNA-specific mapping parameters to optimize sequence yield have been explored (Schubert et al. 2012, Martiniano et al. 2020, Poullet and Orlando 2020, Oliva et al. 2021).”
BackgroundaMeta: an accurate and memory-efficient ancient metagenomic profiling workflow
Pochon, Z., Bergfeldt, N., Kırdök, E., Vicente, M., Naidoo, T., van der Valk, T., Altınışık, N. E., Krzewińska, M., Dalén, L., Götherström, A., et al. (2023) · Genome biology · doi:10.1186/s13059-023-03083-9 · Cited by 74 · cited 1×
“…and quality, including post-mortem DNA damage patterns (Green et al. 2010, J � onsson et al. 2013, Skoglund et al. 2014), fragment length distributions (Green et al. 2010), the breadth of genome coverage (Pochon et al. 2023), and sequence read mapping biasses (van der Valk et al. 2021).”
BackgroundSelf-citationSystematic benchmark of ancient DNA read mapping
Oliva, A., Tobler, R., Cooper, A., Llamas, B., & Souilmi, Y. (2021) · Briefings in Bioinformatics · doi:10.1093/bib/bbab076 · Cited by 63 · cited 3×
“Presently, the most common mapping tools for aDNA reads are BWA-aln (also known as BWA-backtrack), Bowtie2, and BWA-mem, each with applied aDNA parameters (Poullet and Orlando 2020, Oliva et al. 2021) and post-filtering map quality (MQ) scores of generally ≥ 25.”
BackgroundMethodAssessing DNA sequence alignment methods for characterizing ancient genomes and methylomes
Poullet, M., & Orlando, L. (2020) · Frontiers in Ecology and Evolution · doi:10.3389/fevo.2020.00105 · Cited by 61 · cited 4×
“3.5.1 (Langmead and Salzberg 2012) on the --sensitive setting in the default end-to-end mode as recommended for ancient samples (Poullet and Orlando2020), or BWA-mem v0.”
BackgroundMethodInfluentialSelf-citationQuantifying and reducing spurious alignments for the analysis of ultra-short ancient DNA sequences
de Filippo, C., Meyer, M., & Prüfer, K. (2018) · BMC Biology · doi:10.1186/s12915-018-0581-9 · Cited by 53 · cited 1×
“To avoid false inferences, aDNA studies often adopt a conservative minimal read length filter of 35 bp, potentially excluding a substantial portion of usable data (de Filippo et al. 2018, van der Valk et al. 2021).”
MethodFeuerborn, T. R., Palkopoulou, E., van der Valk, T., von Seth, J., Munters, A. R., Pečnerová, P., Dehasque, M., Ureña, I., Ersmark, E., Lagerholm, V. K., et al. (2020) · BMC Genomics · doi:10.1186/s12864-020-07229-y · Cited by 44 · cited 1×
“1) with a human genome as a decoy (Feuerborn et al. 2020), Black rhi-noceros (GCA_020826845.”
BackgroundSelf-citationMeisner, J., Albrechtsen, A., & Hanghøj, K. (2021) · BMC Bioinformatics · doi:10.1186/s12859-021-04375-2 · Cited by 30
Xu, W., Lin, Y., Zhao, K., Li, H., Tian, Y., Ngatia, J. N., Ma, Y., Sahu, S. K., Guo, H., Guo, X., et al. (2021) · Ecology and Evolution · doi:10.1002/ece3.7056 · Cited by 20 · cited 1×
“7.17 (Li 2013) with ancient parameters (-t 8 -k 19 -r 2.5) (Xu et al., 2021).”
Influential
Cited by
Reference genome choice compromises population genetic analyses
Akopyan, M., Genchev, M., Armstrong, E. E., & Mooney, J. A. (2025) · Cell · doi:10.1016/j.cell.2025.08.034 · Cited by 57
Ancient RNA expression profiles from the extinct woolly mammoth
Mármol‐Sánchez, E., Fromm, B., Oskolkov, N., Pochon, Z., Dehasque, M., Aslanzadeh, M., Bozlak, E., Brown, K., van der Valk, T., Kalogeropoulos, P., et al. (2025) · Cell · doi:10.1016/j.cell.2025.10.025 · Cited by 17
“While aDNA studies rely on tools optimized for short, damaged sequences 31 (e.g., Bowtie2 32 and BWA 33 ), methods for aRNA alignment remain less explored, though Bowtie2 32 and Bowtie 34 have been used.”
Self-citationA High‐Throughput Ancient DNA Extraction Method for Large‐Scale Sample Screening
Gilardet, A., Lord, E., Oteo‐García, G., Xenikoudakis, G., Douka, K., Wooller, M. J., Rowe, T. B., Martin, M. D., Moullec, M. L., Anisimov, M., et al. (2025) · Molecular Ecology Resources · doi:10.1111/1755-0998.14077 · Cited by 12
“We investigated mapping bias and length distribution with the AMBER program (Dolenz et al. 2024) on dedupli-cated and indel-realigned BAM files with a generated MD tag (samtools calmd).”
Self-citationOn the ancestry and evolution of the extinct dire wolf
Gedman, G. L., Morrill, K. M., Oppenheimer, J., Hyseni, C., Cassatt‐Johnstone, M., Lucaci, A. G., Alexandre, N. M., Troy, W., Chao, J. C., Fedrigo, O. D., et al. (2026) · Cell Genomics · doi:10.1016/j.xgen.2026.101306 · Cited by 14
“In parallel, we used established methods 21,28–30 to reconstruct the mitochondrial pa-leogenome at high coverage (average depth = 158.4 × ).”
InfluentialSelf-citationGünther, T., Goldberg, A., & Schraiber, J. G. (2025) · G3 Genes Genomes Genetics · doi:10.1093/g3journal/jkaf172 · Cited by 11
Metadata
- Authors
- Dolenz, S., van der Valk, T., Jin, C., Oppenheimer, J., Sharif, M. B., Orlando, L., Shapiro, B., Dalén, L., & Heintzman, P. D.
- Affiliations
- Stockholm University (SE)
- Science for Life Laboratory (SE)
- University of California, Santa Cruz (US)
- Centre National de la Recherche Scientifique (FR)
- Howard Hughes Medical Institute (US)
- Corresponding author
- Dolenz, Stephanie
- van der Valk, Tom
- Heintzman, Peter D
- ORCID
- Venue
- Bioinformatics
- Publisher
- Oxford University Press (OUP)
- Published
- 2024-07-01
- Volume
- 40
- Issue
- 7
- Pages
- btae436
- ISSN
- 1367-4803 (ISSN-L)
- 1367-4811
- PMC
- PMC11254355
- OpenAlex
- W4400344131
- Semantic Scholar
- 5286f2e13052f5fce206274e9091ad8577d3c838
- Cited by
- 33
- References
- 28
- Citations by year
- 2026: 19 · 2025: 13 · 2024: 1
- FWCI
- 4.48
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Bioinformatics (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- HAL (submittedVersion, cc-by)
- eScholarship (California Digital Library) (submittedVersion, cc-by)
- Funders
- Knut and Alice Wallenberg Foundation (2021.0048, KAW 2022.0033)
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Phylogenetic Studies
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Forensic and Genetic Research
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Genetic diversity and population structure
- Keywords
- reference bias, ancient DNA, read mapping, read length, amber, cytosine deamination, fragment size distribution, contamination, paleogenomics, reference genome selection
- MeSH terms
- DNA, Ancient, Algorithms, Animals, Humans, Software, Sequence Alignment, Sequence Analysis, DNA, Computational Biology
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.