The “FORAGESELECT” collaborative project aims to develop and implement Genome Wide Selection (GWS) in breeding programs for perennial ryegrass (Lolium perenne L.). The goal is to improve the genetic response in important agronomic traits, such as forage and seed yield and resistance to biotic and abiotic stresses.
The study is performed on 995 F2 families originated from the DLF breeding program. All families were genotyped by reduced representation sequencing. A total of 1,020,065 SNPs were detected and used for genomic prediction.

First analyses, used for model testing, have been carried out on salt stress tolerance. Ryegrass families where sown in rockwool blocks (in four replicates) in greenhouse, and allowed to establish over 60 days using standard fertilization and watering. Three consecutive treatments, with increasing salt (NaCl) concentrations, were applied. Ten days after initiation of each treatment, the percentage of green matter was evaluated by visual scoring and by digital imaging.

Preliminary analysis using GBLUP have identified a significant amount of genetic variance (individual heritabilities ranging between 0.20 and 0.40 and family heritabilities up to about 0.15). Genomic information was able to explain nearly all the genetic variance. Reliability of genomic predictions for salt resistance was evaluated by cross validation, and showed an increase of 60% compared with the reliability of a single phenotypic record. Further analyses will be performed also on forage and seed yield and rust resistance. SNPs models and Bayesian techniques will also be tested.