Jove
Visualize
Contact Us

Related Experiment Videos

Data mining of the GAW14 simulated data using rough set theory and tree-based methods.

Liang-Ying Wei1, Cheng-Lung Huang, Chien-Hsiun Chen

  • 1Institute of Biomedical Sciences, Academia Sinica, Huafan University, Taipei, Taiwan. ejohn@ibms.sinica.edu.tw

BMC Genetics
|February 3, 2006
PubMed
Summary

This study used rough set theory and decision trees to find disease susceptibility genes. While the first stage accurately predicted traits, neither method identified the true disease-related genes.

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 Novel Genetic Risk Variants Associated With Early-Onset Ischemic Stroke in Taiwan.

Neurology·2026
Same author

The Taiwan Precision Medicine Initiative provides a cohort for large-scale studies.

Nature·2025
Same author

Population-specific polygenic risk scores for people of Han Chinese ancestry.

Nature·2025
Same author

Assessing spatial variability in observed infectious disease spread in a prospective time-space series.

International journal of health geographics·2025
Same author

Multi-ancestry polygenic risk scores for the prediction of type 2 diabetes and complications in diverse ancestries.

medRxiv : the preprint server for health sciences·2025
Same author

Dietary Habits, TCM Constitutions, and Obesity: Investigating the Protective Effects of Vegetarian Dietary Patterns in Taiwan.

Healthcare (Basel, Switzerland)·2025
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

Area of Science:

  • Genetics
  • Data Mining
  • Bioinformatics

Background:

  • Complex datasets in industrial processes often contain vagueness and uncertainty.
  • Genetic Analysis Workshop 14 simulated data involved four layers: disease-related loci, endophenotypes, phenotypes, and a disease trait.
  • Correlations between genetic layers can be obscured, making direct detection of susceptibility genes challenging.

Purpose of the Study:

  • To propose and evaluate a two-stage approach using rough set theory and decision trees for identifying genes associated with a disease trait.
  • To assess the effectiveness of these data mining techniques in uncovering hidden genetic patterns.

Main Methods:

  • A two-stage process was employed: decision trees were built in the first stage to predict trait values using phenotypes.

Related Experiment Videos

  • Phenotypes from the first stage were used in the second stage with rough set theory to find minimal gene subsets linked to the disease trait.
  • Decision trees were also used in the second stage for comparison to map susceptible genes.

    • Main Results:

    • The decision trees in the first stage achieved approximately 99% accuracy in predicting the disease trait.
    • Both decision trees and rough set theory, when applied in the second stage, were unsuccessful in identifying the true disease-related loci.

    Conclusions:

    • The proposed two-stage method, while effective for trait prediction, did not successfully pinpoint the causative genes in this simulated dataset.
    • Further refinement of data mining techniques is needed for accurate identification of susceptibility genes in complex genetic architectures.