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

Adaptive double self-organizing maps for clustering gene expression profiles.

H Ressom1, D Wang, P Natarajan

  • 1Department of Electrical and Computer Engineering, University of Maine, 201 Barrows Hall, Orono, ME 04469-5708, USA.

Neural Networks : the Official Journal of the International Neural Network Society
|July 10, 2003
PubMed
Summary

The adaptive double self-organizing map (ADSOM) offers flexible data partitioning and cluster visualization without prior knowledge. It automatically detects the number of clusters using a novel index, reducing human error in data analysis.

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

Identification of two Escherichia coli pseudouridine synthases that show multisite specificity for 23S RNA.

Biochemistry·1998
Same author

Chemosignal transduction in the vomeronasal organ of garter snakes: cloning of a gene encoding adenylate cyclase from the vomeronasal organ of garter snakes.

Archives of biochemistry and biophysics·1998
Same author

Inhibitory activity of unsaturated fatty acids and anacardic acids toward soluble tissue factor-factor VIIa complex.

Journal of natural products·1998
Same author

A new member of the mouse prolactin (PRL)-like protein-C subfamily, PRL-like protein-C alpha: structure and expression.

Endocrinology·1998
Same author

Pharmacokinetics and biodistribution of radioimmunoconjugates of anti-CD19 antibody and single-chain Fv for treatment of human B-cell malignancy.

Cancer immunology, immunotherapy : CII·1998
Same author

Role of the alpha2-integrin in osteoblast-specific gene expression and activation of the Osf2 transcription factor.

The Journal of biological chemistry·1998

Area of Science:

  • Machine Learning
  • Data Mining
  • Computational Biology

Background:

  • Self-organizing maps (SOMs) are widely used for dimensionality reduction and clustering.
  • Determining the optimal number of clusters often requires a priori knowledge or manual inspection.
  • Existing methods may lack flexibility in topology or automated cluster number detection.

Purpose of the Study:

  • Introduce the adaptive double self-organizing map (ADSOM) for simultaneous data partitioning and cluster visualization.
  • Develop a method for automated cluster number detection without prior information.
  • Evaluate ADSOM's consistency and performance on real-world biological data.

Main Methods:

  • Developed ADSOM by integrating a flexible topology SOM with 2D position vectors.

Related Experiment Videos

  • Implemented a novel index based on hierarchical clustering of position vector locations for automated cluster counting.
  • Tested ADSOM's data partitioning consistency by varying initial node numbers and analyzing common cluster profiles.
  • Applied ADSOM to yeast gene expression data for validation.

    • Main Results:

    • ADSOM successfully performs data partitioning and visualization simultaneously.
    • The novel index enables automated and reliable detection of the number of clusters.
    • ADSOM demonstrates consistent data partitioning across different initial node configurations.
    • Application to yeast gene expression data validates ADSOM's effectiveness.

    Conclusions:

    • ADSOM provides a robust and automated approach to cluster analysis.
    • The method reduces reliance on human interpretation for determining cluster numbers.
    • ADSOM offers a valuable tool for analyzing complex datasets, particularly in bioinformatics.