EUVIMEDEUVIMED entdecken
Open Access

Heterogeneous latent transfer learning in Gaussian graphical models

Qiong Wu iD, Chi Wang iD, Yong Chen iD

Biometrics · 2024

DOI: 10.1093/biomtc/ujae096 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biometrics, 2024. Verfasst von Qiong Wu und 2 weiteren Autor:innen. Beteiligte Einrichtung: Perelman School of Medicine, The University of Pennsylvania, Philadelphia, PA, 19104, United States. Hauptthemen laut MEDLINE: Algorithms, Breast Neoplasms, Computer Simulation, Models, Statistical. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIA NIH HHS, NLM NIH HHS, NIAID NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Gaussian graphical models (GGMs) are useful for understanding the complex relationships between biological entities. Transfer learning can improve the estimation of GGMs in a target dataset by incorporating relevant information from related source studies. However, biomedical research often involves intrinsic and latent heterogeneity within a study, such as heterogeneous subpopulations. This heterogeneity can make it difficult to identify informative source studies or lead to negative transfer if the source study is improperly used. To address this challenge, we developed a heterogeneous latent transfer learning (Latent-TL) approach that accounts for both within-sample and between-sample heterogeneity. The idea behind this approach is to “learn from the alike” by leveraging the similarities between source and target GGMs within each subpopulation. The Latent-TL algorithm simultaneously identifies common subpopulation structures among samples and facilitates the learning of target GGMs using source samples from the same subpopulation. Through extensive simulations and real data application, we have shown that the proposed method outperforms single-site learning and standard transfer learning that ignores the latent structures. We have also demonstrated the applicability of the proposed algorithm in characterizing gene co-expression networks in breast cancer patients, where the inferred genetic networks identified many biologically meaningful gene–gene interactions.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

AlgorithmsBreast NeoplasmsComputer SimulationModels, StatisticalNormal DistributionHumansFemaleMachine LearningGene Regulatory Networks

Stichwörter der Autor:innen

Gaussian graphical modellatent subpopulationprecision matrixtransfer learning

Autor:innen und Institutionen

  • Perelman School of Medicine, The University of Pennsylvania, Philadelphia, PA, 19104, United States.
  • The Center for Health AI and Synthesis of Evidence (CHASE), The University of Pennsylvania, Philadelphia, PA, 19104, United States.
  • Department of Biostatistics, The University of Pittsburgh, Philadelphia, PA, 15261, United States.
  • Division of Cancer Biostatistics, Department of Internal Medicine, College of Medicine and Department of Statistics, The University of Kentucky, Lexington, KY, 40536, United States.

ORCID-Profile: Qiong Wu, Chi Wang, Yong Chen

Förderung

  • NIA NIH HHS
  • NLM NIH HHS
  • NIAID NIH HHS
  • NCATS NIH HHS
  • NIMH NIH HHS
  • NEI NIH HHS
  • NIH HHS
  • National Institutes of Health

Zitieren

APA

Wu, Q., Wang, C., & Chen, Y. (2024). Heterogeneous latent transfer learning in Gaussian graphical models. Biometrics. https://doi.org/10.1093/biomtc/ujae096

Vancouver

Wu Q, Wang C, Chen Y. Heterogeneous latent transfer learning in Gaussian graphical models. Biometrics. 2024. doi:10.1093/biomtc/ujae096

BibTeX
@article{Wu2024,
  author = {Wu, Qiong and Wang, Chi and Chen, Yong},
  title = {Heterogeneous latent transfer learning in Gaussian graphical models},
  journal = {Biometrics},
  year = {2024},
  doi = {10.1093/biomtc/ujae096}
}
Kostenlos starten →