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

Technical note: Computing strategies in genome-wide selection.

A Legarra1, I Misztal

  • 1Institut National de la Recherche Agronomique, UR631 Station d'Amélioration Génétique des Animaux, BP 52627, 32326 Castanet-Tolosan, France. andres.legarra@toulouse.inra.fr

Journal of Dairy Science
|December 22, 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

Computational and modeling approaches for US threshold genetic evaluations of calving ease.

JDS communications·2026
Same author

Approximating reliabilities of indirect predictions using single nucleotide polymorphism effects from large single-step genomic best linear unbiased predictor evaluations.

Journal of dairy science·2025
Same author

Scalar methods to deregress and split genomic predictions, and associated behavior of simple regressions, for later use in combined prediction and validations.

Journal of dairy science·2025
Same author

Efficient implementation of multitrait random regression test-day models with external information for dairy cattle genomic evaluations.

Journal of dairy science·2025
Same author

Study of Genetic Progress in the Context of Disconnection Between Two Originally Connected Populations.

Journal of animal breeding and genetics = Zeitschrift fur Tierzuchtung und Zuchtungsbiologie·2025
Same author

Review: Genomic selection in the era of phenotyping based on digital images.

Animal : an international journal of animal bioscience·2025
Same journal

Protease activities and casein proteolysis in raw and pasteurized bovine milk under neutral and acidic conditions.

Journal of dairy science·2026
Same journal

High-altitude adaptation and associated mortality risk in relocated Holstein cattle.

Journal of dairy science·2026
Same journal

Noninvasive and minimally invasive approaches for acute and chronic stress assessment in dairy cattle.

Journal of dairy science·2026
Same journal

Reference intervals and variability of urinary traits in lactating Holstein cows.

Journal of dairy science·2026
Same journal

The milk fat globule membrane increases bacterial serotonin and exopolysaccharide production for maintenance of intestinal cell differentiation.

Journal of dairy science·2026
Same journal

Effects of rumen-protected methionine and n-3 fatty acid-enriched calcium-salts on biomarkers of liver function in periparturient dairy cows.

Journal of dairy science·2026
See all related articles

Genome-wide genetic evaluations require significant computing power. Preconditioned conjugate gradients and Gauss-Seidel with residuals update offer the fastest solutions for mixed-model equations, saving substantial time and resources.

Area of Science:

  • Quantitative Genetics
  • Bioinformatics
  • Computational Biology

Background:

  • Genome-wide genetic evaluations utilize Best Linear Unbiased Prediction (BLUP)-like estimations.
  • These evaluations involve numerous covariates, such as single-nucleotide polymorphism (SNP) markers, leading to dense mixed-model equations.
  • Solving these equations demands substantial computational resources, including time and storage, even for moderate datasets.

Purpose of the Study:

  • To evaluate and compare the efficiency of various computational strategies for solving dense Henderson's mixed-model equations.
  • To identify optimal algorithms for genetic evaluations involving a large number of markers and records.
  • To assess the performance of matrix-free iterative methods against traditional matrix-based approaches.

Main Methods:

Related Experiment Videos

  • Comparison of computational options including Cholesky decomposition, half-stored Gauss-Seidel, and three matrix-free strategies (Gauss-Seidel, Gauss-Seidel with residuals update, preconditioned conjugate gradients).
  • Implementation and testing of algorithms on a real mouse dataset with 1,928 records and 10,946 SNP markers.
  • Analysis of computing times and resource utilization for each tested method.

Main Results:

  • Preconditioned conjugate gradients and Gauss-Seidel with residuals update achieved solutions in minutes.
  • Half-stored Gauss-Seidel took over an hour, Cholesky decomposition took 2 hours, and matrix-free Gauss-Seidel took 4 days.
  • Gauss-Seidel with residuals update demonstrated efficiency by adjusting residuals, avoiding repeated updates of the equation's left-hand side.

Conclusions:

  • Preconditioned conjugate gradients is the fastest algorithm for solving these dense mixed-model equations.
  • Gauss-Seidel with residuals update is a highly efficient method, particularly suitable for variance component estimation and solving.
  • Matrix-free iterative methods, especially Gauss-Seidel with residuals update, offer significant computational advantages for large-scale genetic evaluations.