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

A balanced accuracy function for epistasis modeling in imbalanced datasets using multifactor dimensionality

Digna R Velez1, Bill C White, Alison A Motsinger

  • 1Center for Human Genetics Research, Vanderbilt University Medical Center, Nashville, Tennessee.

Genetic Epidemiology
|February 27, 2007
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

Predictive value of common genetic variants in idiopathic pulmonary fibrosis survival.

Journal of molecular medicine (Berlin, Germany)·2022
Same author

Genetic heterogeneity: Challenges, impacts, and methods through an associative lens.

Genetic epidemiology·2022
Same author

Identifying genetic markers enriched by brain imaging endophenotypes in Alzheimer's disease.

BMC medical genomics·2022
Same author

Benchmarking AutoML frameworks for disease prediction using medical claims.

BioData mining·2022
Same author

Semantic variation operators for multidimensional genetic programming.

Genetic and Evolutionary Computation Conference : [proceedings]. Genetic and Evolutionary Computation Conference·2022
Same author

Novel digital approaches to the assessment of problematic opioid use.

BioData mining·2022
Same journal

Applying Bayesian Multivariable Mendelian Randomisation to Prioritise Candidate Causal Traits From High-Dimensional Data: Illustration From Estimation of the Effect of Maternal Metabolites on Offspring Birthweight.

Genetic epidemiology·2026
Same journal

Individualized Bayesian Inference Identifies Novel Genetic Variants for Parkinson's Disease.

Genetic epidemiology·2026
Same journal

DRIVE v3: Command Line Application for Identity-by-Descent Haplotype Clustering in Large Biobank Scale Data.

Genetic epidemiology·2026
Same journal

Deep Unsupervised Domain Adaptation for Translating Cancer Dependency Maps From Cell Lines to Breast Cancer Tumor Genomics.

Genetic epidemiology·2026
Same journal

Polygenic Risk Scores for Incident Dementia in the Multi-Ethnic Study of Atherosclerosis.

Genetic epidemiology·2026
Same journal

Outcome and Exposure Polygenic Risk Scores Can Help Reduce Information Bias and Selection Bias in Regression Estimates From Biobank Data.

Genetic epidemiology·2026
See all related articles

Balanced accuracy with an adjusted threshold is the best strategy for detecting epistasis in imbalanced datasets using multifactor dimensionality reduction (MDR). This method outperformed over-sampling and under-sampling techniques, fully recovering detection power.

Area of Science:

  • Genetics
  • Bioinformatics
  • Statistical Genetics

Background:

  • Multifactor dimensionality reduction (MDR) is used to detect epistasis, but struggles with imbalanced datasets.
  • Standard machine learning models may be biased towards the majority class in imbalanced scenarios.

Purpose of the Study:

  • To evaluate strategies for enhancing MDR's power to detect epistasis in imbalanced datasets.
  • To compare over-sampling, under-sampling, and balanced accuracy as fitness functions.

Main Methods:

  • Simulated datasets with varying heritability, minor allele frequency, sample size, and case-control ratios were generated.
  • Three strategies were tested: over-sampling, under-sampling, and balanced accuracy (with/without adjusted threshold).
  • Performance was evaluated based on the ability to detect two-locus epistatic interactions.

Related Experiment Videos

Main Results:

  • Balanced accuracy with an adjusted threshold significantly outperformed both over-sampling and under-sampling.
  • The balanced accuracy approach fully recovered the power to detect epistasis.
  • Over-sampling and under-sampling were less effective in imbalanced datasets.

Conclusions:

  • Balanced accuracy with an adjusted threshold is recommended for MDR analysis of epistasis in imbalanced datasets.
  • This approach provides a more robust and powerful method for genetic interaction detection.
  • Standard accuracy as a fitness measure should be replaced by balanced accuracy in such analyses.