Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Consensus and Meta-analysis regulatory networks for combining multiple microarray gene expression datasets.

Emma Steele1, Allan Tucker

  • 1Centre for Intelligent Data Analysis, Department of Information Systems and Computing, Brunel University, Kingston Lane, Uxbridge Middlesex UB8 3PH, UK. emma.steele@brunel.ac.uk

Journal of Biomedical Informatics
|March 14, 2008
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Optimal Set of Features for Leukaemia Images with Extracted Areas of Interest.

Studies in health technology and informatics·2025
Same author

Extracting Regions of Interest and Selective Feature Application in Leukaemia Image Classification.

Studies in health technology and informatics·2025
Same author

Weight trajectories in aging humanized APOE mice with translational validity to human Alzheimer's risk population: A retrospective analysis.

PloS one·2025
Same author

Identifying and handling data bias within primary healthcare data using synthetic data generators.

Heliyon·2024
Same author

Harnessing Large-Scale Herbarium Image Datasets Through Representation Learning.

Frontiers in plant science·2022
Same author

New JBI policy emphasizes clinically-meaningful novel machine learning methods.

Journal of biomedical informatics·2022
Same journal

Evaluation of temporal preservation in synthetic longitudinal patient data.

Journal of biomedical informatics·2026
Same journal

ARKE: An ontology-driven framework for automated mapping of local radiology procedure terms to the LOINC-RadLex playbook using large language model.

Journal of biomedical informatics·2026
Same journal

A validation-driven training controller for cross-lingual biomedical NER via reinforcement learning-based adaptive loss weighting.

Journal of biomedical informatics·2026
Same journal

ASP-HR: An Adaptive Spatial Perception and Hierarchical Reasoning mechanism for document-level biomedical relation extraction.

Journal of biomedical informatics·2026
Same journal

Beyond Accuracy: Safety-Centered guidelines for the evaluation of LLM-based therapy recommendation systems for chronic multimorbidity patients.

Journal of biomedical informatics·2026
Same journal

DeepEN: A deep reinforcement learning framework for personalized enteral nutrition in critical care.

Journal of biomedical informatics·2026
See all related articles

Combining multiple microarray datasets improves gene regulatory network models. Novel post-learning aggregation methods, Meta-analysis and Consensus Bayesian networks, outperform single-dataset or pre-learning approaches for robust network inference.

Area of Science:

  • Systems Biology
  • Bioinformatics
  • Computational Biology

Background:

  • Microarray data is crucial for modeling gene regulatory interactions from expression levels.
  • Increasingly, multiple microarray datasets are available, offering opportunities for more robust network models.
  • Directly combining diverse microarray datasets is challenging due to experimental biases and platform differences.

Purpose of the Study:

  • To compare pre- and post-learning aggregation frameworks for combining microarray datasets to model gene regulatory networks.
  • To introduce and evaluate two novel post-learning aggregation approaches: Meta-analysis Bayesian networks and Consensus Bayesian networks.

Main Methods:

  • Compared pre-learning aggregation (e.g., scale-normalization before dataset concatenation) with post-learning aggregation (combining models from individual datasets).

Related Experiment Videos

  • Developed Meta-analysis Bayesian networks by aggregating statistical confidences of network edges across datasets.
  • Developed Consensus Bayesian networks by identifying consistent network features across multiple datasets.

    • Main Results:

    • Both novel post-learning aggregation methods improved regulatory network models compared to single-dataset approaches.
    • The proposed methods also outperformed models derived from a simple scale-normalized aggregated dataset.
    • Validation was performed on synthetic and real (Escherichia coli, yeast) gene expression datasets.

    Conclusions:

    • Post-learning aggregation strategies, particularly Meta-analysis and Consensus Bayesian networks, offer superior performance for building gene regulatory networks from multiple microarray datasets.
    • These novel methods effectively handle biases inherent in combining diverse experimental data.
    • The findings suggest these approaches enhance the robustness and reliability of inferred regulatory networks.