Genomic Selection Course

Overview

The Genomic Selection in Breeding course is an intensive, in-person training designed for plant breeding scientists, graduate students, postdocs, and professionals who want to implement genomic selection (GS) in real breeding programs using R. Genomic selection uses genome-wide markers to predict breeding values for polygenic traits and is now widely adopted in public and private programs. This 11th edition focuses on the practical deployment of GS: from basic quantitative genetics to mixed models, G×E modeling, cross-validation, and genomic mating. The course combines concise theory with hands-on labs and a problem-based learning approach. You will work directly in R/RStudio with real datasets, scripts, and workflows that you can adapt to your own program.

Learning outcomes

• Build genomic selection models from scratch and understand key statistical concepts.
• Fit and interpret linear mixed models (BLUEs/BLUPs) for breeding data.
• Apply G×E models, multi-trait analysis, and hybrid prediction.
• Explore optimal parental contributions and genomic mating designs for multi-criteria breeding.

Target audience & prerequisites

• Plant breeders, quantitative geneticists, applied statisticians in breeding companies or public programs.
• Graduate students and postdocs working with phenotypic & genotypic data.
• Basic knowledge of plant breeding and fundamentals of statistics is expected.
• Some familiarity with R is helpful but not strictly required.

Participants who complete the course and associated work will receive a certificate of attendance (equivalent to 2.5 ECTS / approx. 5 academic credits).

Program (Tentative)

The course runs from Monday 18 May to Friday 22 May 2026. Teaching hours are 09:00–17:00 (Mon–Thu) and 09:00–13:00 (Fri), with coffee breaks and a lunch break.

Day 1 — Mon, 18 May: Quantitative Genetics & R

• Review of quantitative genetics and QTL.
• Sources of quantitative trait variation.
• Breeding values, heritabilities, and response to selection.
• Resemblance among relatives; pedigree vs. kinship matrix.
• R/RStudio warm-up and working with breeding data.

Day 2 — Tue, 19 May: GS Introduction & Linear Models

• Big picture of genomic selection and machine-learning perspective.
• Key factors affecting GS and training set optimization.
• Linear models, ANOVA, and GLM in R.
• Fixed vs. random effects; BLUEs and basic mixed-model concepts.

Day 3 — Wed, 20 May: Statistical Concepts for GS

• BLUP vs. BLUE in breeding trials.
• Mixed models and statistical frameworks for GS.
• Construction of kinship/relationship matrices.
• Imputation strategies and marker data preparation.

Day 4 — Thu, 21 May: GS Models & GxE

• One-step and two-step GS models.
• Cross-validation schemes tailored to breeding programs.
• Hybrid prediction and use of testers.
• Modeling genotype × environment interaction (GxE).

Day 5 — Fri, 22 May: Future Directions & Genomic Mating

• GxE (continued) and simulation ideas.
• Genomic mating plans and optimal parental contributions.
• Discussion of case studies and participants’ own data.
• Wrap-up, Q&A, and next steps.

Note: The program may be adjusted slightly to better match participants’ backgrounds and interests.