Featured Writer: Mei Ling Wong, on behalf of the Spring Wheat Breeding Program, Montana State University
Last summer, my advisor Dr. Jason Cook, trained three of his students: Jared Lile, Zoey Pasternak, and me on selecting new experimental lines out of the F5 generation plant populations. We used “Vida,” a high- yielding, widely grown variety, as a reference for overall plant characteristics, and the much older variety “Fortuna” as a benchmark for plant height. We selected the individual plots with desirable traits, including high tiller number, erect growth habit, disease resistance, and large wheat heads. Thinking I could outdo the other students, I tried to be the early bird that caught the best-looking worms. I arrived earlier at the field than my colleagues and selected all the best-looking wheat, hoping they would be the high yielding and high-quality ones. After the growing season, when we looked at the test weight and protein content for the lines we picked, none of the lines I chose were among the top three performers in terms of grain yield, test weight and protein content. Turns out, my “breeder’s eye” was not as sharp as I thought.
Visual selection, while a fundamental tool of plant breeding, cannot always reveal the “hidden potential” in plants. For example, traits like gluten strength and baking quality are not visible to our naked eyes. Because of complex and time-consuming laboratory procedures, often these end-use quality traits are not measured until later stages of the breeding pipeline. This means we have no end-use quality information for lines in earlier generations, limiting the ability to select for these traits in the early stages of breeding. So, what other tools can breeders use to identify superior lines for end-use quality traits?
One of the tools called “genomic selection” has been adapted in many plant breeding programs for improving specific traits, such as grain yield and quality in earlier generations. This tool requires two pieces of key information: first, identify genetic variation in plants’ DNA markers; second, phenotypic data, measuring the plant’s observable traits, such as height, yield, and end-use quality. With this data, statistical models can be developed to link the relationships between genetic markers and the phenotypic traits. This information allows us to estimate the performance of phenotypically untested individuals based only on their genomic data. By selecting individuals with higher predicted values, we can increase genetic gain and improve the efficiency of breeding programs.
This new breeding technology is now being implemented by Jared Lile, a PhD candidate in Plant Genetics and Breeding, who is exploring the use of historical breeding data to develop predictive models for the end-use quality of current spring wheat lines. He aims to identify genetic markers associated with specific quality traits and apply this knowledge to future breeding efforts. Ultimately, we hope to develop new varieties that better meet the needs of the milling and baking industries and help satisfy their customers. This will help maintain Montana’s hard red spring wheat varieties reputation as being renowned for their high-protein content and end-use quality characteristics.