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

Clustering gene expression data using adaptive double self-organizing map.

Habtom Ressom1, Dali Wang, Padma Natarajan

  • 1Intelligent Systems Laboratory, Department of Electrical and Computer Engineering, University of Maine, Orono, Maine 04469, USA. ressom@eece.maine.edu <ressom@eece.maine.edu>

Physiological Genomics
|April 4, 2003
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

Establishment and Characterization of an Immortalized Porcine Satellite Cell Line from China Junmu No.1 Pigs.

Veterinary sciences·2026
Same author

Janus nanofiber membrane integrates exudate management and immunomodulation for enhanced diabetic wound healing.

Colloids and surfaces. B, Biointerfaces·2026
Same author

Salinity Modulates Hormetic Responses of the Estuarine Diatom <i>Chaetoceros muelleri</i> to Short-Chain Chlorinated Paraffins: Insights from Physiology and Transcriptomics.

Environmental science & technology·2026
Same author

Comparative toxicity and molecular recognition of galaxolide and its phototransformation product galaxolide lactone in <i>Daphnia magna</i>.

Environmental science. Processes & impacts·2026
Same author

Transformation and transport: Polyvinyl chloride microplastics modulate fipronil accumulation and toxicity in zebrafish.

Journal of environmental sciences (China)·2026
Same author

Biomimetic PD-1 macrophage nanoplatform for Cip2a silencing and immune activation in oral squamous cell carcinoma.

Journal of nanobiotechnology·2026
Same journal

Regulation of RGS2 is an important piece to the puzzle of syncytiotrophoblast dysfunction in preeclampsia.

Physiological genomics·2026
Same journal

Maternal Estradiol Impacts Fetal HPA Axis and Pulmonary Transcriptomes in Sheep.

Physiological genomics·2026
Same journal

OBESITY AS AN EPIGENETIC CONTINUUM: DEVELOPMENTAL ROOTS, ADIPOSE REMODELING, AND THE LIMITS OF REVERSIBILITY.

Physiological genomics·2026
Same journal

<i>Akkermansia muciniphila</i>: a member of the gut microbiota that modulates intestinal stem cell function.

Physiological genomics·2026
Same journal

Differential gene transcription following intravenous injection of air bubbles in rats with varying resistance to decompression sickness.

Physiological genomics·2026
Same journal

Genomic informational field theory to identify genetic associations of a complex trait using a small sample size.

Physiological genomics·2026
See all related articles

This study introduces adaptive double self-organizing map (ADSOM), a novel clustering technique that automatically determines the number of clusters without prior knowledge. ADSOM enhances data analysis by integrating clustering and visualization for biological data.

Area of Science:

  • Computational biology
  • Machine learning
  • Data mining

Background:

  • Clustering algorithms often require a predefined number of clusters.
  • Existing methods like Double Self-Organizing Map (DSOM) have limitations in parameter control and convergence.
  • Accurate cluster number identification is crucial for interpreting complex biological datasets.

Purpose of the Study:

  • To present Adaptive Double Self-Organizing Map (ADSOM), a novel clustering technique.
  • To enable simultaneous clustering and visualization without a priori knowledge of cluster numbers.
  • To automate the detection of the optimal number of clusters using a novel index.

Main Methods:

  • Development of ADSOM based on DSOM with adaptive parameter updates.
  • Utilizing two-dimensional position vectors for simultaneous clustering and visualization.

Related Experiment Videos

  • Introduction of a novel index based on hierarchical clustering for automated cluster number detection.
  • Application to gene expression data from yeast, human, and mouse.

    • Main Results:

    • ADSOM demonstrates flexible topology and simultaneous clustering and visualization.
    • The technique converges to a consistent number of clusters with appropriate initial node count.
    • Automated cluster detection via the novel index reduces human error compared to visual inspection.
    • ADSOM's performance in identifying cluster numbers is validated against a model-based clustering method.

    Conclusions:

    • ADSOM offers a robust and automated approach for clustering and determining the number of clusters.
    • The method is effective for analyzing gene expression data across different biological systems.
    • ADSOM provides a valuable tool for researchers needing to analyze data with an unknown number of clusters.