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

Cell and tumor classification using gene expression data: construction of forests.

Heping Zhang1, Chang-Yung Yu, Burton Singer

  • 1Department of Epidemiology and Public Health, Yale University School of Medicine, New Haven, CT 06520-8034, USA. heping.zhang@yale.edu

Proceedings of the National Academy of Sciences of the United States of America
|March 19, 2003
PubMed
Summary

Deterministic forests improve cell and tumor classification using gene expression data. This method offers reproducible and interpretable results, outperforming single classification trees.

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

Metapopulations, the Inflationary Effect, and Consequences for Public Health.

The American naturalist·2025
Same author

Adjunctive Probiotic <i>Lactobacillus rhamnosus</i> Probio-M9 Administration Enhances the Effect of Anti-PD-1 Antitumor Therapy <i>via</i> Restoring Antibiotic-Disrupted Gut Microbiota.

Frontiers in immunology·2021
Same author

Sex-specific associations between traumatic experiences and resting-state functional connectivity in the Philadelphia Neurodevelopmental Cohort.

JCPP advances·2021
Same author

Brain Regions Identified as Being Associated with Verbal Reasoning through the Use of Imaging Regression via Internal Variation.

Journal of the American Statistical Association·2021
Same author

Genomics landscape of 185 Streptococcus thermophilus and identification of fermentation biomarkers.

Food research international (Ottawa, Ont.)·2021
Same author

iProbiotics: a machine learning platform for rapid identification of probiotic properties from whole-genome primary sequences.

Briefings in bioinformatics·2021

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Gene chips and microarray data offer powerful tools for cell, tumor, and cancer classification.
  • Recursive partitioning trees are a common methodology for analyzing gene expression data.

Purpose of the Study:

  • To introduce and evaluate a deterministic procedure for creating forests of classification trees.
  • To enhance classification and prediction accuracy in tumor and cell classification using gene expression data.

Main Methods:

  • Exploiting and expanding recursive partitioning tree methodology.
  • Developing a deterministic procedure to form forests of classification trees.
  • Comparing the performance of deterministic forests with random forests and single trees using leave-one-out procedures on two published datasets.

Related Experiment Videos

Main Results:

  • Deterministic forests demonstrate performance comparable to random forests in terms of error rates.
  • Both deterministic and random forests significantly outperform single classification trees.
  • Graphical presentations aid in the interpretation of complex forest structures.

Conclusions:

  • Deterministic forests provide a reproducible and scientifically interpretable alternative for classification tasks using gene expression data.
  • The developed methodology offers numerical improvements and enhanced interpretability for cancer and cell classification.
  • This approach facilitates comparison with existing biological literature and advances the application of gene expression data in diagnostics.