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 Variational Bayes Genomic-Enabled Prediction Model with Genotype × Environment Interaction.

Osval A Montesinos-López1, Abelardo Montesinos-López2, José Crossa3

  • 1Facultad de Telemática, Universidad de Colima, 28040, México.

G3 (Bethesda, Md.)
|April 10, 2017
PubMed
Summary

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

Statistics and data science unlock the predictive power of quantitative genetics.

Frontiers in plant science·2026
Same author

Multimodal genomic prediction is not a buzzword: why modern plant breeding must integrate genomics, enviromics, and phenomics.

G3 (Bethesda, Md.)·2026
Same author

Genomic language model-based genomic prediction in plant breeding.

Trends in plant science·2026
Same author

Comparing statistical 'phenomic prediction' models for remote-sensing-based phenotyping of maize susceptibility to common rust.

Plant phenomics (Washington, D.C.)·2026
Same author

Correction: Multi-trait and multi-environment genomic prediction enhances yield components improvement in durum wheat.

Frontiers in plant science·2026
Same author

Pharmacological treatment patterns, factors associated with glycemic control, and renal function parameters in a real-world cohort of Hispanic adults with type 2 diabetes.

Biomedical reports·2026

A new variational Bayes genomic model accelerates predictions of genotype-by-environment (G×E) interactions. This method is 10x faster than standard Bayesian models, aiding genotype selection in complex agricultural settings.

Area of Science:

  • Genomics
  • Statistical Genetics
  • Machine Learning

Background:

  • Genomic models incorporating genotype-by-environment (G×E) interactions are crucial for crop improvement.
  • Traditional Bayesian models face computational challenges with large datasets, while non-Bayesian models struggle with convergence.
  • Variational Bayes (VB) offers a faster alternative to Markov Chain Monte Carlo (MCMC) methods in machine learning.

Purpose of the Study:

  • To develop a computationally efficient genomic variational Bayes (GVB) model for G×E interactions.
  • To implement a GVB model using half-t priors for robust and non-informative posterior inferences.
  • To evaluate the performance of the proposed GVB model against a standard Bayesian G×E model.

Main Methods:

  • Proposed a novel genomic variational Bayes (GVB) approach for G×E interactions.
Keywords:
GenPredGenomic SelectionShared Data Resourcesgenome-enabled predictionmulti-environmentvariational Bayes

Related Experiment Videos

  • Utilized half-t priors on standard deviation terms for non-informative posterior inferences.
  • Derived and implemented full conditional and variational posterior distributions theoretically.
  • Compared GVB with a standard Bayesian G×E model using maize and wheat datasets.

    • Main Results:

    • The GVB model achieved prediction accuracies comparable to, though slightly lower than, the standard Bayesian model.
    • The GVB model demonstrated a significant computational advantage, being approximately 10 times faster.
    • Inferences from the GVB model were robust and not sensitive to hyper-parameter choices.

    Conclusions:

    • The proposed genomic variational Bayes model provides a computationally efficient alternative for analyzing G×E interactions.
    • This method is particularly valuable for researchers needing to predict and select genotypes across multiple environments.
    • The GVB approach balances prediction accuracy with substantial gains in computational speed.