Speeding Wheat Variety Development | Case Study

Accelerating the development of superior wheat varieties

GEMS multi-trait predictor enables wheat breeders to enrich for varieties with many favorable traits simultaneously, bringing improved varieties to market years ahead of schedule.

Light Grey

Impact and partners

The University of Minnesota’s wheat breeding team, currently headed by Jim Anderson, has been breeding elite materials for farmers to use commercially throughout the state and country since 1888. Three wheat varieties from the program have made the top-20 list of all genetic agricultural products created at the university.

The Minnesota Wheat Research and Promotion Council was created in 1975, and funds numerous programs in the state that directly improve wheat growing practices. They also fund lobbyists in the state capitol and Washington DC. In 2022 they funded this collaborative project between GEMS and the University of Minnesota’s wheat breeding team to translate methods tested in barley over to wheat.

Process and problem solved

University wheat breeders typically take 9 years to complete the cycle, which starts with the crossing of two parental varieties and results in commercial varieties improved in many traits. With rapidly changing climatic conditions and related crop pest pressures, finished varieties may be obsolete soon after introduction. Even a reduction of 2 years in the cycle can make a big difference in impact and commercial viability.

GEMS Services Used in Speeding Wheat Variety Development

GEMS Weather
API

GEMS Soils
API

GEMS Crop Calendar API

GEMS Multi-trait Predictor API

View the GEMS Exchange APIs

The ability to predict crop traits in new environments in silico

For any crop there is continual pressure to produce new varieties to combat the continual emergence of new pests (insects, viruses, fungi and bacteria) and abiotic threats (e.g., drought, salinity, heat stress). Also, it is crucially important to maintain sufficient grain yield while achieving desired quality goals. Quality goals may include considerations such as high protein content for bread baking, increased straw strength so plants don’t get knocked down in storms, altering heading dates so grain harvest can be adjusted to the locale of interest, or keeping height low so it won’t fall over and can support heavier grain heads.

GEMS has developed a multi-trait prediction algorithm that, in principle, should be applicable to almost any non-clonal crop with sufficient genetic variation. The elements required include:

5000 genetic markers consistently recorded across at least 100-150 non-duplicative varieties of a single crop.
Growth trial data for at least 3 years at 15 geographically distinct sites for those varieties.
Phenotypic scoring for the above 3 x 15 environments, marking how well the varieties performed across the variables of interest (e.g., yield, protein content, straw strength).

Results

Anticipated reduction of the wheat breeding cycle by about 2 years.
Several traits (grain yield, protein content, straw strength, heading date, height, and disease resistance) are simultaneously being improved each cycle.
Provide tools to wheat breeders to allow them to see how any variety will perform in different geographic locations across the state.

Summary

We have extensively tested the algorithm in barley and found that it was able to predict important variables (yield, protein content, plant height, heading date, test weight) with correlations of predicted value vs. actual observation of between 0.87 - 0.95. This was true of both the predicted performance for the founding population that had already been tried in the field in years prior, and for the offspring that had never been tested anywhere. Buoyed by the success in barley, we have received funding from the Minnesota Wheat Research and Promotion Council to generalize the prediction tool to wheat, with the goal of greatly improving the efficiency of the university’s wheat breeding process.