Diagnosing scientific replicability through probabilistic distinguishability
Bioinformatics · 03.05.2026 · 42 (5)
DOI: 10.1093/bioinformatics/btag140 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 23.03.2026. Verfasst von Peng Wang und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Mathematics, Jilin University, Changchun, Jilin 130012, China. Hauptthemen laut MEDLINE: Computational Biology, Software. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIGMS NIH HHS, NIH HHS, NSF.
Abstract (Originalsprache)
Motivation Despite the widely recognized importance of replicability in biological research, computational methods to quantify irreplicability and identify irreplicable instances remain underdeveloped. This article presents an efficient and robust computational framework to address this gap. Results To tackle the challenge of defining an acceptable level of intrinsic heterogeneity among replicable studies, we introduce a distinguishability criterion, ensuring that replicable effects, while potentially heterogeneous, can be distinguished from zero effects and maintain consistent directions with high probability. We implement a Bayesian model criticism approach, reporting a Bayesian P-value to identify potential irreplicable instances. Through numerical experiments, we demonstrate the efficacy of the proposed methods in detecting batch effects in high-throughput experiments and identifying instances of the publication bias. Finally, we apply the framework to multi-tissue eQTL data from the GTEx consortium, uncovering tissue-specific eQTLs that represent biological heterogeneity across tissues. Availability and implementation An R package DiscRep implementing our method is available on GitHub (https://github.com/PengWang96/DiscRep).
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- School of Mathematics, Jilin University, Changchun, Jilin 130012, China.
- Department of Statistics and Data Science, Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi, P.O. Box 7909, United Arab Emirates.
- Department of Statistics, Florida State University, Tallahassee, FL 32306, United States.
- Department of Biostatistics, University of Michigan, Ann Arbor, MI 48109, United States.
ORCID-Profile: Peng Wang, Xiaoquan Wen
Förderung
- NIGMS NIH HHS
- NIH HHS
- NSF
- NIH
Zitieren
APA
Wang, P., Cao, H., & Wen, X. (2026). Diagnosing scientific replicability through probabilistic distinguishability. Bioinformatics, 42(5). https://doi.org/10.1093/bioinformatics/btag140
Vancouver
Wang P, Cao H, Wen X. Diagnosing scientific replicability through probabilistic distinguishability. Bioinformatics. 2026;42(5). doi:10.1093/bioinformatics/btag140
BibTeX
@article{Wang2026,
author = {Wang, Peng and Cao, Hongyuan and Wen, Xiaoquan},
title = {Diagnosing scientific replicability through probabilistic distinguishability},
journal = {Bioinformatics},
year = {2026},
volume = {42},
number = {5},
doi = {10.1093/bioinformatics/btag140}
}
EUVIMED