Genome-assisted vine breeding programs

Breeding schemes usually involve first choosing parents and then selecting offspring within crosses. Genomic prediction is suitable both for predicting cross mean and for classifying genotypes within a cross. These steps correspond to the components of the predictive ability of genomic prediction. The cross mean is the sum of the breeding values ​​of the parents if the allelic effects are only additive, but in practice some deviation may result from dominance or epistasis. To date, only a few studies have investigated the predictive ability of cross-mean in heterozygous cultures, and none have investigated the parameters that influence it.

Very few authors have assessed the potential usefulness of genomic prediction in grapevine (Vitis vinifera subsp. viniferous). Before genomic selection can be used in grapevine, the predictive ability must be assessed across populations. In particular, predictive ability can be assessed using a diversity panel and biparental offspring as training and validation sets, respectively. This is a difficult configuration, given the low genetic relatedness, but this configuration is much more likely to occur in actual selection schemes than genomic prediction within the same population. As with grapes, studies looking at genomic prediction across the population are also lacking for most clonally propagated crops.

Recently, scientists from IFV-INRAE-Institut Agro tested cross-population genomic prediction in a more realistic breeding configuration. They evaluated the quality of the prediction on 15 traits of interest (related to yield, berry composition, phenology and vigour) both for the mean genetic value of each cross (average cross) and the genetic values ​​of individuals within each cross (individual values). Genomic prediction under these conditions proved useful: for the cross-mean, the mean predictive ability per trait was 0.6. The predictive ability per crossover was halved on average, but peaked at 0.7. The average predictive ability of the individual values ​​within the crosses was 0.26, or about half of the intra-dialle value taken as a reference. For some traits and crosses, these predictive ability values ​​across the population are promising for the implementation of genomic selection in grapevine breeding.

“We implemented genomic prediction in grapevine in a selection context, i.e. across populations, on 15 traits, in ten related crosses, and obtained moderate to high PA values ​​for certain crosses and traits, thus showing the usefulness of genomic prediction in grapevine. Never before has genomic prediction been implemented for so many traits and crosses simultaneously in this species,” said Dr Agnès Doligez. These results will greatly help in designing genomic prediction-assisted grapevine breeding programs.




Charlotte Brault1,2,3Vincent Segura1.2Patrice This1.2Loic Le Cunff1,2,3Timothée Flutre4Pierre Francois1.2Thierry Pons1.2Jean-Pierre Peros1.2Agnes Doligez1,2,*


1 Geno-Vine UMT, IFV-INRAE-Institut Agro, F-34398 Montpellier, France

2 UMR AGAP Institute, Univ Montpellier, CIRAD, INRAE, Agro Institute, F-34398 Montpellier, France

3 French Institute of Vine and Wine, F-34398 Montpellier, France

4 Paris-Saclay University, INRAE, CNRS, AgroParisTech, GQE-Le Moulon, 91190, Gif-sur-Yvette, France

About Dr. Agnès Doligez

Dr. Agnès Doligez is currently working on the Genetic Improvement and Adaptation of Mediterranean and Tropical Plants (AGAP), National Institute for Agronomic Research (INRA). Dr. Doligez conducts research in agricultural plant science and genetics, and their current project is “Development of genetic resources and tools to breed new cultivars better adapted to the problems of global warming”.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of press releases posted on EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.