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

Bayesian learning of sparse gene regulatory networks.

Zeke S H Chan1, Lesley Collins, N Kasabov

  • 1Knowledge Engineering and Discovery Research Institute (KEDRI), Auckland University of Technology, Auckland, New Zealand. shun.chan@aut.ac.nz

Bio Systems
|January 16, 2007
PubMed
Summary

This study introduces sparse Bayesian learning (SBL) to simplify gene regulatory network (GRN) models. SBL reduces model complexity, improving the accuracy and interpretability of inferred gene relationships.

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

Genomic Scans across Three Eucalypts Suggest that Adaptation to Aridity is a Genome-Wide Phenomenon.

Genome biology and evolution·2017
Same author

Evolving fuzzy neural networks for supervised/unsupervised online knowledge-based learning.

IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2008
Same author

Proceedings of the SMBE Tri-National Young Investigators' Workshop 2005. Investigating the intron recognition mechanism in eukaryotes.

Molecular biology and evolution·2005
Same author

Fast neural network ensemble learning via negative-correlation data correction.

IEEE transactions on neural networks·2005
Same author

A hybrid genetic algorithm and expectation maximization method for global gene trajectory clustering.

Journal of bioinformatics and computational biology·2005
Same author

Complex spliceosomal organization ancestral to extant eukaryotes.

Molecular biology and evolution·2005

Area of Science:

  • Systems Biology
  • Computational Biology
  • Bioinformatics

Background:

  • Differential equations (DEs) are widely used for modeling gene regulatory networks (GRNs).
  • DE models often suffer from over-parameterization, especially for large numbers of genes (d), leading to data over-fitting and difficult interpretation.
  • The O(d^2) parameter requirement in DEs poses a significant challenge for complex GRN inference.

Purpose of the Study:

  • To address the over-parameterization issue in DE-based GRN modeling.
  • To introduce and apply sparse Bayesian learning (SBL) for GRN model sparsification.
  • To enhance the plausibility, interpretability, and consistency of inferred GRNs.

Main Methods:

  • Application of sparse Bayesian learning (SBL) to sparsify GRN models represented by differential equations.

Related Experiment Videos

  • Leveraging the parsimony principle to drive redundant parameters to zero.
  • Validation using time-series gene expression data from yeast Saccharomyces cerevisiae.

    • Main Results:

    • Successfully sparsified GRN models using SBL, reducing the number of effective parameters.
    • Inferred GRNs were more plausible and interpretable due to the sparse parameter sets.
    • Achieved more optimal and consistent solutions by reducing the solution space volume.

    Conclusions:

    • Sparse Bayesian learning is an effective method for overcoming over-parameterization in differential equation-based GRN models.
    • SBL enhances the biological relevance and interpretability of inferred gene regulatory networks.
    • The approach reliably reproduced known regulatory events, demonstrating its practical utility.