Pith. sign in

REVIEW 5 major objections 6 minor 18 references

Emerging Challenges in Molecular Paleontology: Misapplication of Environmental DNA Fragments and Misconception of Deamination as a Key Criterion for In Situ DNA Identification

T0 review · 5 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper argues that standard ancient-DNA identification—built on probe enrichment, target-genome BLAST, and deamination filtering—can mistake environmental DNA for the fossil's own DNA, and that the Tianyuan Cave Man genome is therefore…

desk verdict A cautionary commentary with a credible general point, but its headline claim about Tianyuan rests on unpublished BLAST hits and a misreading of deamination. read the letter →

arxiv 2412.06378 v1 pith:WZL4OULK submitted 2024-12-09 q-bio.GN

Wan-Qian Zhao , Shu-Jie Zhang , Zhan-Yong Guo , Zeng-Yuan Tian , Gang-Qiang Cao , Mei-Jun Li , Li-You Qiu , Jin-Yu Yang
show 4 more authors
Yong-Kai Wang Shu-Hui Zhang Zhi-Fang Zheng Min-Zhi Wu
This is my paper · ORCID
classification q-bio.GN
keywords ancientDNAenvironmentalcontaminationdeaminationBLASTE-valueTianyuanCaveManLycopterafossilsinsitufossiltaphonomy
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This review challenges the assumption that DNA recovered from a fossil is the original DNA of the single individual to which the fossil belongs. The authors argue that fossils are porous rock bodies continuously flushed by water, so environmental DNA from later periods can repeatedly enter and replace the original DNA. They use their own screens of 120-million-year-old Lycoptera fish fossils to claim that fragments matching ape and chimpanzee genomes, with no detectable deamination damage, are ancient hominin DNA that infiltrated the rock. From this they conclude that the assembled genome of the so-called Tianyuan Cave Man is likely a mixture of unclassified Homo species and modern humans, and that nearly all studies built on the standard method deserve reevaluation. If correct, many published ancient-human genome interpretations would need to be revisited.

What carries the argument

The argument is carried by two instruments: a sequence-uniqueness screen using BLAST against all known genomes, requiring the target species to be the top hit with an E-value below 1e-07 and an E-value gap of more than 1e-02 over the second-best match, and a physical model of fossils as porous rocks that continuously exchange old and new environmental DNA through wet-dry cycles, named 'less old and more new molecules'. The authors apply this screen to Lycoptera fossil data and interpret the resulting Hominoidea- and Pan-matching fragments as ancient hominin DNA, which they then use to argue that the standard method's reliance on deamination as an authenticity marker is misplaced.

What would settle it

Re-run the Lycoptera data against an updated all-genome database and check whether the 22 Pan-unique fragments still top-hit Pan with an E-value gap greater than 1e-02; then attempt independent retrieval of the same fragments from fresh Lycoptera powder in a dedicated clean lab, and test whether identical fragments also appear in surrounding sediments, groundwater, or modern reagents. If the fragments do not reproduce or also appear in modern water, the ancient-hominin interpretation fails.

Watch

Extended reading notes

Core claim

The central claim is that fossil bones behave as open, water-flushed rock: environmental DNA continuously enters internal cavities after burial, so the DNA recovered from a fossil is not necessarily the original DNA of the fossil's species. On this basis the authors argue that the widely used ancient-DNA workflow mislabels foreign fragments as endogenous: probe capture at 62–65°C is not species-specific for fragments longer than about 30 base pairs; BLAST searches restricted to the target species' genome admit environmental lookalikes; and deamination filters select damaged fragments with high E-values while excluding well-preserved original DNA. The authors report that Lycoptera fossils yielded more than 230,000 fragments hitting Hominoidea genomes and 22 fragments uniquely matching the Pan genome, none with detectable deamination, and they take these as evidence that hominin DNA has been seeping into fossils since the fossils formed. The conclusion follows: the so-called Tianyuan Cave Man genome is likely a mix of DNA from multiple Homo species, modern humans, or even other genera, and the findings of studies using the same method should be reevaluated.

Load-bearing premise

The inference that fish-fossil DNA fragments matching ape and chimpanzee genomes are ancient hominin DNA depends on assuming that BLAST top-hit ranking with the stated E-value criteria reflects biological origin, that absence of detectable deamination rules out modern contamination, and that the fragments are not conserved sequences shared across taxa or alignment artifacts.

Editorial extensions

If this is right

  • If the central claim is right, published ancient-human genomes assembled with the standard capture-and-deamination workflow cannot be treated as representing single individuals.
  • Ancient-DNA reference databases should be expanded from the target species to all known genomes before any fragment is assigned to a species.
  • Fragments that fail the proposed uniqueness criteria, including those with E-value gaps smaller than 1e-02, should be discarded before genome assembly.
  • Deamination should not be used as a positive marker of ancient origin, and heavily deaminated fragments should be avoided because they produce inflated E-values and unreliable BLAST hits.
  • The Tianyuan Cave Man mitogenome and chromosome 21 assembly are specifically called into question as potentially mixed with DNA from many other species.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A direct test of the contamination hypothesis would be to re-analyze the Tianyuan and other published ancient-DNA datasets with the proposed all-genome BLAST screen: if most assigned fragments fail the uniqueness criteria, the paper's central worry is confirmed; if they remain uniquely assigned, the worry is at least bounded.
  • The permeability argument implies that even perfectly clean laboratory conditions cannot certify in-situ origin, shifting the burden of proof to depositional context and taphonomic sealing.
  • If deamination is not a necessary criterion, some claims of extremely old DNA may be under-estimated rather than over-estimated, because well-preserved sealed fossils would be excluded by the standard filter.
  • The same 'less old and more new molecules' mechanism suggests that sediment, rock voids, and petroleum reservoirs near fossil sites are promising sources of ancient DNA, extending the paper's proposal beyond fossils proper.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. The paper is a critical methodology review of ancient DNA (aDNA) research following the approach of Pääbo's team. It argues that fossil bones are open systems into which environmental DNA (eDNA) continuously infiltrates, that probe enrichment and BLAST-based screening are insufficient to exclude eDNA, and that deamination is not a reliable criterion for identifying original in situ DNA. The central empirical claim, developed in the section 'Fossils Contain Many DNA Fragments from Ex Situ Species,' is that DNA fragments from 120-million-year-old Lycoptera fossils that BLAST-match Pan genomes, while lacking deamination, must originate from ancient Homininae and therefore that the assembled 'Tianyuan Cave Man' genome is likely a mixture of genomes from different genera and periods, with the findings of nearly all studies using Pääbo's method requiring reevaluation. The paper also proposes stricter BLAST criteria, including an E-value below 1E-07 and a 1E-02 gap to the second hit, for species-unique assignment.

Significance. If the central empirical claims were established, the paper would have considerable significance for molecular paleontology, because it would imply widespread contamination in published ancient human genomes. The paper does identify legitimate concerns: fossils are porous, eDNA can enter bone over geological time, deamination is not a universal or sufficient marker of ancient origin, and short DNA fragments are intrinsically difficult to assign taxonomically. However, the manuscript's own load-bearing evidence is not demonstrated: the key Lycoptera fragment analyses are confined to the authors' unpublished preprints (refs 11 and 12), no negative controls or quantitative validations are presented, and the inference from BLAST top-hit rankings to 'ancient Homininae eDNA' is unsupported. The paper also contains no reproducible code or accessible data beyond a request-based offer, and its proposed thresholds are introduced without calibration. The deamination critique is partially grounded in Lindahl's work and deserves consideration, but the broad claim that nearly all Pääbo-method studies require reevaluation is disproportionate to the evidence supplied.

major comments (5)
  1. [§3 (Fossils Contain Many DNA Fragments from Ex Situ Species), pp. 3-4 and Figure 1] The identification of the 22 'Pan-unique' Lycoptera fragments as ancient hominin eDNA is not established. The fragments are short, can have Identity or Cover below 90%, and show no detectable deamination, yet the paper concludes that they 'must originate from ancient Homo species or close relatives' solely from BLAST top-hit ranking with an E-value gap of 1E-02 to the second hit. Short queries can top-hit a particular genome through conserved sequences, low-complexity regions, or alignment artifacts, and BLAST E-values do not by themselves determine biological origin. The manuscript provides no negative controls, no mapping of the fragments back to Lycoptera or fish genomes, no demonstration of hominin-specific substitutions, and no assessment of whether the fragments could be alignment artifacts. Since this inference is the sole empirical basis for the Tianyuan-mixture conclusion, the central claim collapses without it.
  2. [§3, references 11 and 12, Table S3B of reference 11] The main empirical evidence for the contamination mechanism is contained in the authors' own unpublished preprints (refs 11 and 12), not in this manuscript. The manuscript states that the 22 fragments are listed in Table S3B of reference 11 and that full sequencing data are 'available upon request,' but no data, code, or stable supplementary material is provided here. This makes the central claim unverifiable and effectively circular: the criteria used to define 'Pan-unique' fragments are established only in those preprints and are then cited as though they validate the present conclusions. The absence of accessible data is a load-bearing reproducibility problem, not a stylistic one.
  3. [§3, p. 4 and §5 ("Deamination" Should Not Be Viewed as an Essential Criterion), pp. 5-6] The absence of detectable deamination cannot serve as evidence that the Lycoptera fragments are ancient Homininae DNA. Modern contamination also lacks deamination, and the paper does not demonstrate a preservation mechanism that would prevent deamination specifically in these fragments. The volcanic-tuff description is asserted for the Lycoptera fossils in general and is not tied to the sequenced fragments. Consequently, the logical chain 'no deamination + not 100% identical to Pan = early Homininae origin' is invalid: sequencing error, missing reference diversity, conserved non-hominin sequence, or recent contamination are equally consistent with the observations.
  4. [§4 (Probe Enrichment Does Not Exclude Sequences from Ex Situ Species), p. 5 and Figure 2] The 15-base sequence CCTATAGCACCCCCTC is used to argue that the 'Tianyuan Cave Man' mitogenome may be mixed with foreign fragments, but a 15-mer that matches many taxa is expected by chance alone, and the figure does not show that such matches occur among the actual sequencing reads or that they influence the assembled mitogenome KC417443. No quantitative comparison (e.g., the number of reads matching non-target taxa, or a test of whether the observed 15-mer is overrepresented in the library) is provided. This evidence therefore cannot support the conclusion that the mitogenome is contaminated.
  5. [§7 (Exploring aDNA Research Methods), p. 7] The proposed BLAST criteria are ad hoc and uncalibrated. The manuscript recommends an E-value below 1E-07 and an E-value difference greater than 1E-02 between the top hit and the second hit, but it provides no derivation, no error model, and no validation on datasets of known composition. These thresholds also do not resolve the central difficulty, because reference databases are incomplete and short fragments can satisfy the criteria while still being non-diagnostic. As presented, the proposed method is a set of rules rather than a tested analytical framework.
minor comments (6)
  1. [References, ref. 14] The author name is misspelled as 'Lindalh' in reference 14; it should be 'Lindahl'.
  2. [References, refs. 3 and 4] Reference titles are truncated (e.g., 'The deep population history of northern...' and 'Ancient genomes revealed the complex hu...'), and citation formatting is inconsistent throughout.
  3. [Throughout, pp. 3-5] The terms 'MS mode,' 'minimal E-value mode,' and 'MS Model' are used without definition in this manuscript; they should be defined or a precise pointer to the method in reference 11 should be given.
  4. [Introduction, p. 3] Informal passages such as 'As one observes the rain this morning nourishing the Cretaceous rocks' and the phrase 'less old and more new molecules' are not appropriate for a journal article and should be rewritten.
  5. [§3, p. 4] The statement that artificially synthesizing and faking the fragments could cost millions of dollars is irrelevant to the scientific argument and should be removed.
  6. [Figure 1 caption] The caption refers to sequence IDs (379-533, 290-378, 534-555) but the manuscript does not provide a table mapping these IDs to sequences; readers should be referred to a stable supplementary file rather than a preprint.

Circularity Check

2 steps flagged · score 5.0 of 10

Central empirical premise (22 'Pan-unique' Lycoptera fragments) is sourced only from the authors' own unpublished preprints, making the Tianyuan contamination conclusion depend on a self-citation chain.

  1. self citation load bearing [pp. 3-4, 'Fossils Contain Many DNA Fragments from Ex Situ Species' (Lycoptera BLAST results and inference)]
    "using the 'MS mode' online, the results indicated that 22 fragments uniquely aligned with the Pan-genome... Notably, there is no detectable 'deamination' (Table S3B of reference 11; full sequencing data available upon request). Therefore, the result indicates that these fragments do not originate from modern Pan-genomes but rather from the genomes of early Homininae species (Figure 1A and 1B)."

    This is the load-bearing inference that eDNA from later hominins infiltrated 120-million-year-old Lycoptera fossils, which then supports the conclusion that the Tianyuan genome is a mixture and that 'nearly all studies employing Pääbo's method warrant a reevaluation.' Every empirical component of this inference — the 22 fragments, the 'MS mode' criterion, the absence of deamination (Table S3B), and the volcanic-tuff classification — is located in reference 11, an arXiv preprint by the same authors, with raw data 'available upon request.' The paper does not reproduce the data or provide an external benchmark; it cites its own unpublished work as the authority for the fact.

  2. uniqueness imported from authors [p. 7, 'Exploring aDNA Research Methods' (proposed uniqueness criteria)]
    "The BLAST results should indicate an E-value for the target species with less than 1E-07 to reduce the risk of systematic identification errors. In addition, if the BLAST hits multiple species, the target one must be ranked highest and have an E-value that differs from the E-value of the second match by more than 1E-02; otherwise, the identification should be deemed inconclusive."

    These thresholds and the 'MS mode'/'minimal E-value mode' used to select the 22 fragments are introduced in the authors' own reference 11 (and 12). The paper then uses fragments selected by this self-authored criterion as evidence that the criterion can identify ancient hominin eDNA. The criterion is thus validated by the same data it was used to produce, and its authority is imported from the authors' prior preprints rather than from an external standard.

full rationale

The paper is not entirely circular: the deamination critique independently cites Lindahl (ref 14), and the demonstration that the single Tianyuan sequence CCTATAGCACCCCCTC matches many taxa is shown in the paper's own Figure 2. However, the central empirical claim about Tianyuan — that the assembled genome is likely a mix of genomes from other genera — rests on the Lycoptera BLAST results, the 'MS mode' selection, the absence of deamination in those fragments, and the volcanic-tuff classification, all of which are delegated to the authors' own arXiv preprints (refs 11 and 12). The paper explicitly says the data are 'available upon request' and directs readers to the Methods in reference 11, so the key evidence is not independently verifiable within the manuscript. The proposed uniqueness criteria are also imported from those same preprints and used to select the fragments that then serve as evidence for the eDNA-infiltration mechanism, creating a self-reinforcing loop. Thus the score reflects partial circularity: significant self-citation is load-bearing for the headline conclusion, while the deamination and probe-enrichment critiques retain some independent content. A score of 5, rather than higher, is appropriate because the paper does contain an independent, in-paper BLAST observation and an external chemical reference for the deamination argument.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new physical entities. Its load-bearing assumptions are about fossil hydrology, BLAST uniqueness, and the meaning of deamination. The E-value and affinity thresholds are hand-chosen criteria rather than fitted parameters, but they function as free choices in the proposed method. The main empirical support is self-cited and not independently verifiable from this paper.

free parameters (2)
  • E-value thresholds for species uniqueness = 1E-07 for target, 1E-02 gap to second hit
    Hand-chosen criteria proposed for the method, not calibrated against known positive and negative controls. The central identification claim depends on these thresholds.
  • Affinity selection window for Pan-genome hits = 38% to 90% (Figure 1B)
    A hand-set filter used to select 266 of the 6,727 Pan-aligning fragments. No justification is given for the boundaries.
assumptions (4)
  • domain assumption Fossils behave as open hydrological systems through interconnected internal spaces.
    The central contamination mechanism. It is introduced through a pore-space calculation in reference 11 and a rainwater thought experiment on page 3, not measured or demonstrated in this paper.
  • domain assumption Deamination requires liquid water and is suppressed in sealed mineral environments.
    Relies on Lindahl (reference 14). Used to argue that absence of deamination does not disprove ancient origin, a key step in the argument.
  • ad hoc to paper A fragment is assignable to a species only if its BLAST E-value is a unique top hit with a gap over 1E-02 to the second hit.
    Proposed as a gold standard without validation on known positive and negative controls. The paper's conclusions depend on this criterion being diagnostic.
  • ad hoc to paper Absence of detectable deamination in Pan-like fragments from Cretaceous fossils proves they are ancient hominin DNA rather than modern contamination.
    This is the load-bearing inference that converts BLAST hits into the contamination claim. It is stated on pages 3-4 without excluding modern human contamination, database artifacts, or conserved sequences.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Emerging Challenges in Molecular Paleontology: Misapplication of Environmental DNA Fragments and Misconception of Deamination as a Key Criterion for In Situ DNA Identification." pith.science (2026). https://pith.science/paper/WZL4OULK

@misc{pith2026241206378,
  author       = {Pith},
  title        = {Pith review of: Emerging Challenges in Molecular Paleontology: Misapplication of Environmental DNA Fragments and Misconception of Deamination as a Key Criterion for In Situ DNA Identification},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WZL4OULK}},
  note         = {Machine review of arXiv:2412.06378}
}
read the original abstract

This article critically examines the methodologies applied in ancient DNA (aDNA) research, particularly those developed by Dr. P\"a\"abo's team, which have significantly influenced the field. The focus is on the challenges of distinguishing original in situ DNA (oriDNA) from environmental DNA (eDNA) contamination in fossil samples. Recent analyses indicate that even with rigorous extraction and sequencing protocols, a considerable amount of eDNA remains present, often misinterpreted as oriDNA. This misidentification risks the accuracy of species ascription and evolutionary interpretations derived from fossil analyses. The paper explores fossil preservation's physical and chemical dynamics, which allow eDNA from similar and disparate species to infiltrate bone matrices. We propose enhancements to methodological frameworks, such as broader BLAST database usage and stringent E-value criteria, to improve species-specific aDNA identification. Additionally, the article critiques the reliance on deamination patterns as a definitive marker for aDNA, suggesting a reevaluation of this criterion due to its inconsistency and the potential for misleading sequencing results. Ultimately, our findings advocate for a more cautious and refined approach to aDNA research, ensuring more reliable and verifiable scientific outcomes

Figures

Figures reproduced from arXiv: 2412.06378 by the authors.

Figure 1
Figure 1. Analysis of 236,544 Sequences Aligned to the Hominidae Family (Apes) [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

18 extracted references · 18 canonical work pages

  1. [1]

    Fu et al., DNA analysis of an early modern human from Tianyuan Cave, China

    Q. Fu et al., DNA analysis of an early modern human from Tianyuan Cave, China. Proceedings of the National Academy of Sciences 110, 2223-2227 (2013)

    work page 2013

  2. [2]

    Dabney et al., Complete mitochondrial genome sequence of a Middle Pleistocene cave bear reconstructed from ultrashort DNA fragments

    J. Dabney et al., Complete mitochondrial genome sequence of a Middle Pleistocene cave bear reconstructed from ultrashort DNA fragments. Proceedings of the National Academy of Sciences 110, 15758-15763 (2013)

    work page 2013

  3. [3]

    Cell 184, 3256-3266.e3213 (2021)

    Xiaowei Mao et al., The deep population history of northern...m the Late Pleistocene to the Holocene. Cell 184, 3256-3266.e3213 (2021)

    work page 2021

  4. [4]

    Bai et al., Ancient genomes revealed the complex hu...ctions of the ancient western Tibetans current biology

    F. Bai et al., Ancient genomes revealed the complex hu...ctions of the ancient western Tibetans current biology. Current Biology 34, 2594-3605.e2597 (2024)

    work page 2024

  5. [5]

    Lackner, N

    M. Lackner, N. Helmbrecht, S. Pääbo, S. Riesenberg, Detection of unintended on - target effects in CRISPR genome editing by DNA donors carrying diagnostic substitutions. Nucleic Acids Research 51, e26-e26 (2023)

    work page 2023

  6. [6]

    Mafessoni et al., A high-coverage Neandertal genome from Chagyrskaya Cave

    F. Mafessoni et al., A high-coverage Neandertal genome from Chagyrskaya Cave. Proceedings of the National Academy of Sciences 117, 15132-15136 (2020)

    work page 2020

  7. [7]

    Devièse et al

    T. Devièse et al. , Compound -specific radiocarbon dating and mitochondrial DNA analysis of the Ple istocene hominin from Salkhit Mongolia. Nature Communications 10 (2019)

    work page 2019

  8. [8]

    Kanton et al., Organoid single-cell genomic atlas uncovers human-specific features of brain development

    S. Kanton et al., Organoid single-cell genomic atlas uncovers human-specific features of brain development. Nature 574, 418-422 (2019)

    work page 2019

Show all 18 references
  1. [9]

    Douka et al

    K. Douka et al. , Age estimates for homini n fossils and the onset of the Upper Palaeolithic at Denisova Cave. Nature 565, 640-644 (2019)

    2019

  2. [10]

    Wang et al., 12 Huma n genetic history on the Tibetan Plateau in the past 5100 years

    H. Wang et al., 12 Huma n genetic history on the Tibetan Plateau in the past 5100 years. Science Advances 9, eadd5582 (2023)

    2023

  3. [11]

    W. Q. Zhao et al., Ancient DNA from 120-Million-Year-Old Lycoptera Fossils Reveals Evolutionary Insights (2024 in arXiv)

    2024

  4. [12]

    W. Q. Zhao et al., DNA Fragments in Crude Oil Reveals Earth's Hidden History. (2024 in arXiv)

    2024

  5. [13]

    R. C. Fleischer, T. Maricic, M. Whitten, S. Pääbo, Multiplexed DNA Sequence Capture of Mitochondrial Genomes Using PCR Products. PLoS ONE 5 (2010)

    2010

  6. [14]

    Lindalh, Instability and decay of the primary structure of DNA

    T. Lindalh, Instability and decay of the primary structure of DNA. Nature 362, 709-715 (1993)

    1993

  7. [15]

    A. M. Bailleul, Z. Li, DNA staining in fossil cells beyond the Quaternary: Reassessment of the evidence and prospects for an improved understanding of DNA preservation in deep time. Earth-Science Reviews 216 (2021)

    2021

  8. [16]

    Zheng, A

    X. Zheng, A. M. Bailleul, Z. Li, X. Wang, Z. Zhou, Nuclear preservation in the cartilage of the Jehol dinosaur Caudipteryx. Communications Biology 4 (2021)

    2021

  9. [17]

    Bomfleur, S

    B. Bomfleur, S. McLoughlin, V. Vajda, Fossilized Nuclei and Chromosomes Reveal 11 180 Million Years of Genomic Stasis in Royal Ferns. Science 343, 1376-1377 (2014)

    2014

  10. [18]

    Zhang et al., Denisovan DNA in Late Pleistocene sediments from Baishiya Karst Cave on the Tibetan Plateau Science 370, 584-587 (2020)

    D. Zhang et al., Denisovan DNA in Late Pleistocene sediments from Baishiya Karst Cave on the Tibetan Plateau Science 370, 584-587 (2020)

    2020

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.