Genomic selection for nitrogen use efficiency in perennial ryegrass ( Lolium perenne L.) under field conditions
Junping Wang, M. Michelle Malmberg, Fan Shi, Shane McGlone, Pieter Badenhorst, Noel Cogan, Kevin F. SmithAbstract
Background
Improving the nitrogen use efficiency (NUE) of pastures has the benefit of reducing costs of production and reducing nitrogen loss to the environment. Genetic variation has been shown to exist for NUE, and hence NUE is a trait for breeding programs.
Methods
In this study, we develop genomic selection methods for NUE in perennial ryegrass through developing high‐throughput sensor‐based phenotyping and genotyping by sequencing (GBS) technologies. NUE of an advanced perennial ryegrass breeding population was screened in a spaced plant field trial which contained 644 genotypes, 3 nitrogen treatment levels (0, 20, and 40 kg ha −1 per application), and 3 replicates. The trial was conducted for 2 years with a total of 6 nitrogen applications. An unmanned aerial system (UAS) equipped with multispectral sensors was deployed weekly over the trial. Approximately 4–5 weeks after nitrogen fertilizer application, 75–675 selected samples were cut for ground truthing. Prediction models for biomass were developed based on spectral and ground truth data and biomass for each plant was computed. Plants were genotyped by GBS transcriptomics.
Results
NUE, defined as biomass production per unit of N application, varied significantly with N application level, season and among genotypes. Moderate broad‐sense heritability (0.61–0.72) for NUE was observed. Genomic prediction accuracies were in the range of 0.3–0.5.
Conclusions
Our results demonstrated that genomic selection for NUE was possible. The genomic prediction developed in these advanced breeding lines may be tested in other genetic backgrounds. The technologies are ready to be extended into other perennial pasture grass species.