Performance Story: High-throughput omics approaches for effective breeding selection of durum wheat quality
Durum wheat is an economically important crop and the source of semolina for the production of pasta, couscous, and various types of baked products. Its market value is largely determined by the end-use quality traits. However, quality tests are labour and cost-intensive and most times, large quantities of samples are needed. Therefore, practising earlier selection on quality traits on a large scale within the breeding program is a challenge.
The advent of advanced genetic and genomic approaches provided a feasible approach to predict quality and make a selection on such a large scale at earlier generations and ultimately release variety with improved wheat quality. With this project, the team aimed to develop high-throughput omics approaches to effectively select key quality traits including protein, low molecular weight glutenin subunits (LMW-GS) and high molecular weight glutenin subunits (HMW-GS), milling performance and pigment for Canadian durum wheat. An efficient protein profiling platform with the matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) approach enables to qualitatively identify these complex glutenin subunits and precisely quantify their phenotypic variations were successfully established and optimized. With this approach, 6 LMW-GS and 4 HMW-GS subunits were identified. This also allowed for phenotyping these complex glutenin subunits in a large scale on a durum wheat nested association mapping (NAM) population. In addition, protein, milling performance and pigments were also evaluated on this NAM population.
With the genotypic dataset, a NAM-based genome-wide association analysis (GWAS) was performed to uncover the complex genetic architecture for these quality traits. GWAS identified 200 marker traits associations (MTAs) on 14 durum wheat chromosomes for these quality traits. These include 16 QTL for protein, 62 QTL for LMW-GS, 39 QTL for HWM-GS, 67 QTL for milling performance and 16 QTL for pigment. Further validation of these markers in the durum breeding program will enable the development of valuable omics tools to rapidly select Canadian durum wheat cultivars with high-quality traits using marker-assisted selection (MAS). In addition, the phenomics platform, large phenotypic and genomics datasets will also allow to develop a genomic selection (GS) and machine learning (ML) or deep learning (DL) based omics platform to predict the performance of these complex quality traits.
Key outcomes include:
A High-throughput (HTP) phenotyping platform to qualitatively and quantitatively analyze functional protein molecules: LMW-GS and HMW-GS were developed.
The complex of LMW-GS and HWM-GS was revealed with 6 subunits for LWM-GS and 4 subunits for HWM-GS.
The genetic architecture for complex quality traits was uncovered with 16 QTL for protein, 62 QTL for LMW-GS, 39 QTL for HWM-GS, 67 QTL for milling performance and 16 QTL for pigment, resulted in a total of 200 marker traits associations (MTAs).
The phenomics platform, large phenotypic and genomics datasets and MTAs will allow to develop an omics platform that incorporates marker-assisted selection (MAS), genomic selection (GS), machine learning (ML) or deep learning (DL) to predict the performance of these complex quality traits and select them in a large scale at early generation.