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

Scoring clustering solutions by their biological relevance.

I Gat-Viks1, R Sharan, R Shamir

  • 1School of Computer Science, Tel Aviv University, Tel Aviv 69978, Israel. iritg@post.tau.ac.il

Bioinformatics (Oxford, England)
|December 12, 2003
PubMed
Summary

We developed a new statistical method to evaluate gene clustering, improving biological interpretation. This approach helps select the best clustering results for gene expression 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

A reference model of circulating hematopoietic stem cells across the lifespan with applications to diagnostics.

Nature medicine·2025
Same author

Symptoms and management of cow's milk allergy: perception and evidence.

Frontiers in allergy·2024
Same author

Circadian clock gene disruption in white blood cells of patients with celiac disease.

Biochimie·2023
Same author

Pedi-R-MAPP: The development of a nutritional awareness tool for use in remote paediatric consultations using a modified Delphi consensus.

Clinical nutrition (Edinburgh, Scotland)·2022
Same author

Feasibility of a truncated surveillance schedule for patients following curative intent treatment for carcinoma of the oral cavity.

International journal of oral and maxillofacial surgery·2021
Same author

ANAT 3.0: a framework for elucidating functional protein subnetworks using graph-theoretic and machine learning approaches.

BMC bioinformatics·2021

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Clustering gene expression data is crucial for functional annotation and motif identification.
  • Evaluating and comparing different clustering algorithms remains a challenge in gene expression analysis.
  • Lack of standardized guidelines hinders the selection of optimal clustering solutions.

Purpose of the Study:

  • To develop a statistically sound method for assessing gene clustering solutions based on prior biological knowledge.
  • To enable comparison of different clustering outcomes and optimization of clustering algorithm parameters.
  • To provide a robust scoring mechanism for evaluating the biological relevance of gene clusters.

Main Methods:

  • A novel method projects biological attribute vectors onto a real line to maximize between-group and within-group variance.

Related Experiment Videos

  • The projected data are scored using a non-parametric analysis of variance (ANOVA) test.
  • Confidence of the score is evaluated, and the method is validated with simulated and yeast cell-cycle gene expression data.

    • Main Results:

    • The developed scoring method outperforms existing measures like separation to homogeneity ratio and silhouette measure.
    • The approach effectively assesses clustering solutions against biological knowledge.
    • Demonstrated utility in evaluating multiple clustering results on real biological data.

    Conclusions:

    • The novel statistical method provides a reliable way to assess and compare gene clustering solutions.
    • This approach aids in selecting biologically meaningful clusters from gene expression data.
    • The method offers a valuable tool for optimizing clustering algorithms and interpreting results.