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

A hierarchical unsupervised growing neural network for clustering gene expression patterns.

J Herrero1, A Valencia, J Dopazo

  • 1Bioinformatics, CNIO, Ctra. Majadahonda-Pozuelo, Km 2, Majadahonda, 28220 Madrid Protein Design Group CNB-CSIC, 28049 Madrid, Spain.

Bioinformatics (Oxford, England)
|March 10, 2001
PubMed
Summary

We introduce the Self-Organising Tree Algorithm (SOTA), a neural network for analyzing gene expression data. SOTA provides robust hierarchical clustering, identifying correlated gene patterns efficiently for large datasets.

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

Advanced maternal age and assisted reproductive technologies: outcomes, genomics, and real-world evidence.

Frontiers in reproductive health·2026
Same author

The Ecology of Benthic Diatom Assemblages in Saline Wetlands of the Ebro Basin, NE Spain.

Microbial ecology·2025
Same author

Aripiprazole as protector against COVID-19 mortality.

Scientific reports·2024
Same author

Barcelona scale for buccophonatory apraxia: Quantitative assessment tool.

Neurologia·2022
Same author

Calcium oxalate monohydrate crystalluria in ethylene glycol poisoning confirmed by scanning electron microscopy.

Clinica chimica acta; international journal of clinical chemistry·2022
Same author

Chromatin network markers of leukemia.

Bioinformatics (Oxford, England)·2020

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Machine Learning

Background:

  • DNA array technologies enable rapid, large-scale gene expression monitoring.
  • Identifying correlated gene expression patterns is crucial for biological insights.
  • Traditional clustering methods can be computationally intensive and less robust.

Purpose of the Study:

  • To present a novel unsupervised neural network approach for gene expression data analysis.
  • To introduce the Self-Organising Tree Algorithm (SOTA) for hierarchical clustering.
  • To offer a robust and efficient method for identifying gene expression patterns.

Main Methods:

  • Utilizing the Self-Organising Tree Algorithm (SOTA), a growing neural network that forms a binary tree topology.
  • Implementing a divisive clustering approach, resolving hierarchical levels from top to bottom.

Related Experiment Videos

  • Employing a randomization-based criterion for statistical support in cluster definition.

    • Main Results:

    • SOTA offers advantages over classical hierarchical clustering, including robustness and accuracy.
    • The algorithm provides a statistical criterion for defining clusters and stopping tree growth.
    • Average gene expression patterns are inherently represented by the algorithm's neurons.
    • SOTA exhibits near-linear runtime, making it suitable for massive datasets.

    Conclusions:

    • SOTA is a powerful and versatile tool for analyzing gene expression data.
    • The method provides statistically supported hierarchical clustering with computational efficiency.
    • SOTA is applicable to any numerical dataset with a computable similarity measure.