Genetic and Environmental Variance Among F2 Families in a Commercial Breeding Program for Perennial Ryegrass (Lolium perenne L.)

In the joint project “FORAGESELECT”, we aim to implement Genome Wide Selection (GWS) in breeding of perennial ryegrass (Lolium perenne L.), in order to increase genetic response in important agronomic traits such as yield, seed production, stress tolerance and disease resistance, while decreasing greenhouse emissions and nitrogen loss. GWS model building includes 1) development of a robust quantitative genotyping method for an outcrossing species, 2) tailoring of multi-locational, multi-annual phenotype data, 3) association analysis and development of prediction models.

As part of (2) the aim of this study was to estimate the genetic and environmental variance in the training set composed of F2 families selected from a ten year breeding period. Variance components were estimated on 1193 of those families, sown in 2001, 2003 and 2005 in five locations around Europe. Families were tested together with commercial varieties used as checks.

The first analyses focused on yield (green and dry matter) and the data were analyzed using a mixed model including fixed effect of experiment, effect of check variety, and random effects within experiment effects to recover interblock information, effects of pedigree (parents), repeated effect of the same family and residual error.

Results showed the presence of a significant genetic variance among the random factors, indicating the existence of a considerably high variance in the commercial population. This will provide good opportunities for future improvement programs based on GWS. Future work will focus on developing association models based on tailored phenotype data and genotype-by-sequencing-derived allele frequencies.