Skip to content
DraconiticResearch
Search Sources API
186,546 studies31 sourcesUpdated 6 min ago
PreprintCC BYbioRxiv

Berkson's Paradox: A Common Statistical Error in Genomics Analyses

  • J. Hum
  • K. Korthauer
  • C. G. de Boer1
  1. 1University of British Columbia
Read full textPDFDOICite

Abstract

Berkson's paradox describes a scenario where a spurious correlation arises between two independent variables within a subset of samples selected based on a third variable which the first two variables cause. The conditions for Berkson's paradox are perfect in genomics due to many variables we study being causally intertwined, common selection criteria being biased either intentionally or not by the variables of interest, and our propensity to look towards correlations for windows into underlying mechanism. Here, we provide examples of cases where spurious correlations are likely to arise in common genomics analyses, and suggest analysis approaches to avoid misinterpreting spurious correlations.

Related

  1. Agronomic characterisation of a common buckwheat germplasm collection unlocks resources for genomics-assisted ideotype breeding
    F. Hess, L. Kronenberg, G. Rimat +5
    bioRxivPreprintCC BYFull textPlant biologybioRxiv
  2. Comparative genome analyses suggest a common blueprint for obligate endoparasitism in Strepsiptera
    R. J. A. Millena, P. B. Frandsen, M. James +3
    bioRxivPreprintCC BY-NCFull textGenetics & genomicsbioRxiv
  3. The role of gene regulatory evolution in weapon diversification: Insights from comparative genomics, chromatin profiling, and functional analyses of BMP signaling in horned beetles
    Phillip L Davidson, Isabel G Manley, Armin P Moczek
    PLOS GeneticsPeer-reviewedCC BYFull textMolecular biology & gene regulationCell biologyPLOSEurope PMC
  4. Common attentional mechanisms, but distinct computational dynamics, across Statistical and Reward Learning
    C. Dolci, T. Verguts, W. J. Louter +2
    bioRxivPreprintCC BYFull textNeurosciencebioRxiv
  5. Genome-wide association study and selective sweep analyses identify loci associated with plant height and spike number in a Chinese micro-core collection of common wheat (Triticum aestivum L.)
    Xinqian Lyu, Dongli Lu, Zhenhao Zhu +4
    Frontiers in Plant SciencePeer-reviewedCC BY 4.0Plant biologyFrontiersCrossref

Cite

BibTeXRIS
BibTeX
@misc{hum2026berksons,
  title = {{Berkson's Paradox: A Common Statistical Error in Genomics Analyses}},
  author = {J. Hum and K. Korthauer and C. G. de Boer},
  howpublished = {bioRxiv},
  year = {2026},
  month = oct,
  doi = {10.64898/2026.10.22.761068},
  url = {https://www.biorxiv.org/content/10.64898/2026.10.22.761068v1},
  note = {CC BY},
}
RIS
TY  - UNPB
TI  - Berkson's Paradox: A Common Statistical Error in Genomics Analyses
AU  - J. Hum
AU  - K. Korthauer
AU  - C. G. de Boer
PY  - 2026
DA  - 2026/10/22
PB  - bioRxiv
DO  - 10.64898/2026.10.22.761068
UR  - https://www.biorxiv.org/content/10.64898/2026.10.22.761068v1
AB  - Berkson's paradox describes a scenario where a spurious correlation arises between two independent variables within a subset of samples selected based on a third variable which the first two variables cause. The conditions for Berkson's paradox are perfect in genomics due to many variables we study being causally intertwined, common selection criteria being biased either intentionally or not by the variables of interest, and our propensity to look towards correlations for windows into underlying mechanism. Here, we provide examples of cases where spurious correlations are likely to arise in common genomics analyses, and suggest analysis approaches to avoid misinterpreting spurious correlations.
ER  -