Joint Bayesian additive regression trees for multiple nonlinear dependency networks
Biometrics · 2025
DOI: 10.1093/biomtc/ujaf158 ↗Kurz gefasst
Fachartikel in der Zeitschrift Biometrics, 2025. Verfasst von Licai Huang und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, TX 77030, United States. Hauptthemen laut MEDLINE: Nonlinear Dynamics, Protein Interaction Maps, Models, Statistical. Die Arbeit nennt 39 Literaturangaben. Gefördert von: National Research Foundation of Korea.
Abstract (Originalsprache)
Identifying protein–protein interaction networks can reveal therapeutic targets in cancer; however, for heterogeneous cancers such as colorectal cancer (CRC), a pooled analysis of the entire dataset may miss subtype-specific mechanisms, whereas separate analyses of each subgroup’s data may reduce the power to identify shared relations. To address this limitation, we propose a hierarchical Bayesian model for the inference of dependency networks that encourages the common selection of edges across subgroups while allowing subtype-specific connections. To allow for nonlinear dependence relations, we rely on Bayesian Additive Regression Trees (BART) to characterize the key mechanisms for each subgroup. Because BART is a flexible model that allows nonlinear effects and interactions, it is more suitable for genomic data than classical models that assume linearity. To connect the subgroups, we place a Markov random field prior on the probability of utilizing a feature in a splitting rule; this allows us to borrow strength across subgroups in identifying shared dependence relations. We illustrate the model using both simulated data and a real data application on the estimation of protein–protein interaction networks across CRC subtypes.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, TX 77030, United States.
- Statistical Innovation Group, AbbVie Inc, North Chicago, IL, 60064, United States.
- Biohealth Data Science, Graduate School of Transdisciplinary Health Science, Yonsei University, Seoul 03722, South Korea.
- Department of Health Informatics and Biostatistics, Graduate School of Public Health, Yonsei University, Seoul 03722, South Korea.
ORCID-Profile: Christine B Peterson, Min Jin Ha
Förderung
- National Research Foundation of Korea
Zitieren
APA
Huang, L., Peterson, C. B., & Ha, M. J. (2025). Joint Bayesian additive regression trees for multiple nonlinear dependency networks. Biometrics. https://doi.org/10.1093/biomtc/ujaf158
Vancouver
Huang L, Peterson CB, Ha MJ. Joint Bayesian additive regression trees for multiple nonlinear dependency networks. Biometrics. 2025. doi:10.1093/biomtc/ujaf158
BibTeX
@article{Huang2025,
author = {Huang, Licai and Peterson, Christine B and Ha, Min Jin},
title = {Joint Bayesian additive regression trees for multiple nonlinear dependency networks},
journal = {Biometrics},
year = {2025},
doi = {10.1093/biomtc/ujaf158}
}
EUVIMED