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

Identifying SNPs predictive of phenotype using random forests.

Alexandre Bureau1, Josée Dupuis, Kathleen Falls

  • 1Department of Human Genetics, Oscient Pharmaceuticals, Waltham, Massachusetts, USA. alexandre.bureau@msp.ulaval.ca

Genetic Epidemiology
|December 14, 2004
PubMed
Summary

Random forests effectively identify single-nucleotide polymorphisms (SNPs) associated with complex diseases. This machine learning method analyzes joint effects of SNPs, improving gene mapping for conditions like asthma.

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

Colocalization of eQTLs With Type 2 Diabetes and Glycemic Traits Using Whole-Genome Sequences in Diverse Populations From the NHLBI Trans-Omics in Precision Medicine (TOPMed) Program.

Diabetes·2026
Same author

Correction: A genome-wide investigation of depression among individuals with and without irritability.

Molecular psychiatry·2026
Same author

The reliability and accuracy of recombination inferred by Shapeit2 duoHMM on whole genome sequence.

bioRxiv : the preprint server for biology·2026
Same author

Plasma metabolomic profiles associated with cardiovascular disease in type 2 diabetes from the Trans-Omics for Precision Medicine (TOPMed) program.

Atherosclerosis·2026
Same author

Whole genome sequencing analysis of over 3500 individuals dementia-free over 85 years old.

Journal of Alzheimer's disease : JAD·2026
Same author

Dynamics of Spinal Fluid Immune Cell Alterations Following Cladribine Tablet Treatment in Multiple Sclerosis.

Neurology(R) neuroimmunology & neuroinflammation·2026

Area of Science:

  • Genetics
  • Computational Biology
  • Bioinformatics

Background:

  • Complex diseases involve multiple genes, making genetic mapping challenging.
  • Traditional association studies struggle with numerous single-nucleotide polymorphisms (SNPs).
  • High-dimensional nonparametric methods offer solutions for analyzing large genetic datasets.

Purpose of the Study:

  • To apply random forests for identifying SNPs predictive of disease phenotype.
  • To extend random forest importance measures to capture joint SNP effects.
  • To evaluate the method's performance in complex disease gene mapping.

Main Methods:

  • Utilized random forests, a machine learning algorithm based on classification trees.
  • Employed bootstrap sampling and out-of-bag error estimation for prediction accuracy.

Related Experiment Videos

  • Quantified SNP importance by measuring misclassification increases upon permutation.
  • Extended importance measures to assess joint effects of SNP pairs.

    • Main Results:

    • Random forests successfully identified important SNPs and SNP pairs in an asthma dataset.
    • High importance values correlated with strong SNP-disease associations.
    • Predictive importance did not always perfectly align with traditional association measures.
    • The method demonstrated robustness across various two-locus disease models and non-associated SNPs.

    Conclusions:

    • Random forests provide a powerful tool for complex disease gene mapping using high-dimensional SNP data.
    • The method effectively captures both individual and joint SNP effects.
    • This approach enhances the identification of disease susceptibility genes, as illustrated in asthma research.