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

Application of machine learning in SNP discovery.

Lakshmi K Matukumalli1, John J Grefenstette, David L Hyten

  • 1Beltsville Agricultural Research Center, Bovine Functional Genomics Laboratory, US Department of Agriculture, ARS, Beltsville, MD 20705, USA. lmatukum@gmu.edu

BMC Bioinformatics
|January 10, 2006
PubMed
Summary

This study introduces a machine learning (ML) method to improve the accuracy of single nucleotide polymorphism (SNP) detection. The ML approach significantly reduces false positives, enhancing SNP discovery efficiency.

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

Pangenome-based structural variant imputation enables large-scale genotype-phenotype studies in dairy cattle.

Nature communications·2026
Same author

Contrasting Selection on GxE and Main Effect Loci for Soybean Grain Yield.

Genetics·2026
Same author

Liver transcriptome dynamics in Holstein cows during the periparturient transition.

Scientific reports·2026
Same author

Transcriptomic and Metabolic Insight Into Flavonoid Biosynthesis Underlying Black and Yellow Seed Coat Color Variation in Soybean (<i>Glycine max</i>).

Plant direct·2026
Same author

β-cyclocitral as a cross-species mediator of abiotic stress signaling: insights and future directions toward crop improvement.

Frontiers in plant science·2026
Same author

Phased-assembly-driven pangenome graphs for structural variant genotyping and complex trait mapping in dairy cattle.

Nature communications·2026

Area of Science:

  • Bioinformatics
  • Genetics
  • Machine Learning

Background:

  • Single nucleotide polymorphisms (SNPs) are a major source of genetic variation.
  • Existing polymorphism detection software like PolyBayes and PolyPhred produce high false positive rates.
  • Machine learning (ML) methods show promise in bioinformatics for various prediction tasks.

Purpose of the Study:

  • To develop and evaluate a machine learning (ML) method to augment existing polymorphism detection software.
  • To improve the accuracy of single nucleotide polymorphism (SNP) prediction.
  • To reduce the number of false positive SNP predictions.

Main Methods:

  • Applied the C4.5 machine learning algorithm to classify candidate SNPs.
  • Utilized human expert decisions for training data.

Related Experiment Videos

  • Trained the ML classifier on 27,275 candidate SNPs from soybean, with PolyBayes analysis.
  • Tested the classifier on 18,390 additional candidate SNPs.

    • Main Results:

    • The ML classifier achieved 97.3% agreement with human experts on test data, compared to 7.8% for PolyBayes.
    • Positive predictive values (PPV) for SNPs increased from 7.8% (PolyBayes) to 84.8% (ML), a 5- to 10-fold improvement.
    • The ML method generated significantly fewer false positives (249) than PolyBayes (16,955).

    Conclusions:

    • A machine learning (ML) method effectively enhances polymorphism detection software accuracy.
    • The ML approach significantly reduces human intervention and increases productivity in SNP discovery.
    • The developed ML framework is adaptable for integration with various polymorphism discovery tools.