Enhancing grapevine breeding efficiency through genomic prediction and selection index.

Area of Science:

• Plant breeding
• Genomics
• Viticulture

Background:

• Grapevine breeding is slow (20-25 years) and needs new traits for climate change adaptation.
• Traditional selection relies on phenotypic data and limited markers, lacking a defined ideotype.

Purpose of the Study:

• To accelerate grapevine breeding using genomic prediction.
• To identify superior individuals in Rosé wine and Cognac breeding programs.
• To define breeding ideotypes collaboratively with industry representatives.

Main Methods:

• Applied genomic prediction (Genomic Best Linear Unbiased Predictor, Least Absolute Shrinkage Selection Operator) to 30 traits in two grapevine breeding programs.
• Evaluated predictive abilities and explored across-population prediction.
• Used multivariate selection index for identifying superior individuals.

Main Results:

• Genomic prediction showed variable predictive abilities (0-0.9) across traits, influenced by heritability and trait characteristics.
• Across-population prediction demonstrated potential for leveraging diverse grapevine populations.
• Superior individuals were successfully identified for both Rosé wine and Cognac programs.

Conclusions:

• Genomic prediction is a powerful tool to accelerate grapevine breeding and enhance climate change adaptation.
• Collaborative ideotype definition with industry ensures breeding program relevance.
• The study provides a framework for efficient selection of elite grapevine genotypes.