henhaobio
Sign in

PLoS ONEPaleontologyPeer-reviewed journal article

{micro}CT Scanning Effects on DNA and a Multi-Step Workflow for Archaeological Petrous Bones

Lumila Paula MenéndeziD, Pierre LuisiiD, María Clara López‐SosaiD, Sergio Monteiro Da Silva, Laura Tabitha BuckiD, Susan C. KuzminskyiD, Mirsha Quinto‐SáncheziD, Anne Le MaîtreiD, Christine ChappardiD, Cassandra RiosiD, Hans GrohiD, Wara Siles, Gustavo Montiel Hernández, Lorena Becerra‐ValdiviaiD, Juan Manuel ArgüellesiD, Bernardo Yáñez, Constanza de la Fuente CastroiD, Camila TamburriniiD, Vivette García DeisteriD, Miguel Ángel Contreras-Sieck, Sebastián PasqualiniiD, Paula NovellinoiD, Candela Acosta MoranoiD, Daniela GuevaraiD, Daniela Alit MansegosaiD, , Nicolás RascovaniD

PDFDOIPreprint
In-depth full-text digest available · Jump to digest ↓

Abstract

The petrous portion of the temporal bone is a key element in human evolutionary studies due to its exceptional preservation of biomolecules and morphological information. However, intensive and often redundant sampling has raised concerns about sustainability and long-term conservation. Here we present the first systematic evaluation of whether micro-Computed Tomography ({micro}CT)--a widely used tool for digital preservation--affects ancient DNA (aDNA) integrity in human petrous bones. We analyzed 93 archaeological samples from Argentina, of which 50 had been scanned using {micro}CT and 43 had not. We compared six molecular parameters, including endogenous content, read length, cytosine deamination patterns and contamination estimates. No statistically significant differences were observed between scanned and unscanned samples across any parameter (Mann-Whitney/Wilcoxon tests, p <0.05). Although mitochondrial contamination was marginally higher in scanned samples (p = 0.051), this was not driven by contamination estimates above the widely accepted 5% threshold for genomic analysis, Moreover, this pattern was not observed when considering nuclear contamination. These results indicate that, under appropriate scanning conditions, {micro}CT imaging does not compromise DNA preservation. Building on this evidence, we propose a sustainable, multi-step workflow that integrates biological profiling, osteobiography, imaging, and compositional pre-screening prior to molecular sampling. This interdisciplinary approach maximizes the scientific information obtained from skeletal collections while minimizing destructive practices, thereby promoting ethical and sustainable research on irreplaceable anthropological remains, and fosters collaboration across research fields.

In-depth summary

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (2 Oct 2025) · CC BY 4.0

Fifty micro-CT-scanned archaeological petrous bones from Argentina did not differ significantly from 43 unscanned ones on six ancient DNA metrics, motivating a multi-step workflow that scans bones before destructive sampling.

Petrous bones are now the main target for ancient DNA (aDNA) sampling and are often destroyed with little digital record of their morphology, while it has been unclear whether X-ray exposure from micro-computed tomography (µCT) damages their DNA. Earlier tests used animal bones or human femora, and the authors argue the dense human petrous needs direct testing; they present this as the first systematic test in archaeological human petrous bones.

Key results

  1. SuggestedUnder the scanning settings used, µCT showed no detectable effect on aDNA. Scanned and unscanned samples did not differ significantly on any of six metrics: endogenous content, mean read length, 5' C→T frequency (half-UDG and non-UDG libraries), mitochondrial DNA (mtDNA) contamination and nuclear contamination.

    50 µCT-scanned vs 43 unscanned petrous samples from the same Argentine sites. Mann-Whitney U / Wilcoxon rank-sum tests in R v4.3.2 gave p > 0.05 for every parameter (Table 3, Figure 3). Read length and C→T patterns gave no sign of strand breaks or extra damage caused by scanning. Nuclear contamination was estimated from the X chromosome, in males only. · 4. Testing the Effects of µCT Scanning on aDNA…; Tables; Figure legends; Abstract

  2. SuggestedmtDNA contamination was marginally higher in scanned samples (p = 0.051, not significant), but almost all estimates stayed under the 5% threshold used for genomic work, and nuclear contamination did not differ between groups.

    ContamMix mtDNA contamination: p = 0.051, with 48 out of 49 estimates reported below 5%. X-chromosome nuclear contamination (males only): p = 0.798, with the highest value (2.3%) in an unscanned sample. The authors attribute the mtDNA trend to sample selection, which they did not test directly. · 4. Testing the Effects of µCT Scanning on aDNA…; Abstract; Figure legends

  3. ShownEndogenous DNA content was similar in both groups, and the highest values came from scanned samples.

    Endogenous content (the percentage of unique reads mapping to the human reference) had p = 0.663. The authors explain the high values in scanned samples by a tendency to scan better-preserved bones, not by any effect of scanning. That explanation was not tested. · 4. Testing the Effects of µCT Scanning on aDNA…; Figure legends

  4. ClaimedProposed five-step workflow: macroscopic evaluation; biological profiling and osteobiography; CT plus surface scanning or photogrammetry; compositional pre-screening; then molecular sampling only for a clear research question, ideally one sampling event for radiocarbon, isotopes and DNA.

    The workflow was not tested. Named pre-screening options: nitrogen content (%N), near-infrared spectroscopy (NIR), X-ray fluorescence (XRF) and zooarchaeology by mass spectrometry (ZooMS). Listed challenges: unequal access to scanners, about $100 per scan as a global average, and sample transport. · 5. Proposed Workflow for a Sustainable Petrous…; 6. Main challenges of imaging studies; 7. Final considerations

  5. SuggestedIn the Allen Ancient DNA Resource, petrous bones supplied aDNA for nearly half of sequenced individuals, and the petrous was usually the only bone sampled.

    In AADR v9 metadata, petrous bones were the aDNA source for 47% of all sequenced individuals, and the only sampled bone in 96% of those cases. The text does not describe how these figures were counted. · 3. A Proposal for a Sustainable Workflow in…

Methods, data and reproducibility

Six standard aDNA metrics were compared between 50 petrous samples scanned on a Skyscan 1176 (Bruker) µCT scanner and 43 unscanned petrous samples from the same Argentine Holocene sites, using Mann-Whitney U tests in R. Extraction, library preparation and sequencing followed Rascovan et al. (2025). The authors also propose a multi-step pre-sampling workflow, which they do not test.

  • Archaeological petrous bone samples from Middle to Late Holocene sites in Argentina (Andes, Central Argentina, Cuyo); scanned and unscanned groups come from the same sites but are not matched pairs · human · n = 93 samples (50 µCT-scanned, 43 unscanned)
  • Allen Ancient DNA Resource (AADR v9) metadata, used to count how often petrous bones were the sampled element · human (described as a compendium of ancient human genomes)
Methods & tools
Skyscan 1176 (Bruker) µCT scanner, ContamMix, X chromosome-based nuclear contamination estimation, half-UDG and non-UDG (uracil-DNA glycosylase) library treatment, R v4.3.2, Mann-Whitney U / Wilcoxon rank-sum test, Allen Ancient DNA Resource (AADR v9), 3D Slicer Preview Release 5.9.0, Inkscape
Reagents & environment
Statistics in R v4.3.2; Figure 2 made with 3D Slicer Preview Release 5.9.0 and Inkscape; µCT scanning parameters given in Table 2.
Reproducibility gaps
Supplementary Table S1 (per-sample metadata and molecular metrics) is said to hold all data needed to reproduce the statistics, but no URL is given. No analysis code, sequence repository or accession is listed; wet-lab methods are cited to Rascovan et al. (2025).
Limitations (6)
  • Scanned and unscanned samples came from the same sites but were not elements from the same individuals, so differences cannot be attributed solely to µCT scanning. · stated by the authors
  • The conclusions apply to the scanning conditions used here (Table 2). Other X-ray methods such as synchrotron scanning can exceed DNA-harmful doses and need stricter precautions. · stated by the authors
  • Adding CT imaging delays studies and raises costs (about $100 per scan as a global average), and access to scanners and funding is unequal across regions and institutions. · stated by the authors
  • How to reconcile conflicting molecular and anatomical data, different timescales and hierarchies of evidence is left unresolved. · stated by the authors
  • The evidence comes from Argentine Holocene samples scanned on one scanner under one protocol, and the text reports no power or equivalence analysis. A non-significant result with 50 vs 43 samples therefore does not rule out a modest scanning effect. · noted in this summary
  • Selection bias toward scanning better-preserved bones is used to explain the higher endogenous content and the p = 0.051 mtDNA contamination trend, but it was not tested and could equally hide a real scanning effect. · noted in this summary

Read this if you sample petrous bones for ancient DNA or curate human skeletal collections and want evidence on whether micro-CT scanning before sampling harms DNA, plus a proposed pre-sampling workflow.

Citations
1
FWCI
3.96
Field percentile
top 9%

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISHipsley '20Immel '16Parker '20Harney '20Ibrahim '22Ferrari '20
64% semantic matchEvolutionary Biology
C. A. Hipsley, R. Aguilar, J. R. Black+1 more·bioRxiv·Jan 22, 2020