Scientists from the Centre for Crop and Food Innovation (CCFI) at Murdoch University have participated in a landmark proof-of-concept research project demonstrating how pangenome-guided breeding strategies can reunite two traits that are usually trade-offs, namely stress tolerance and yield.
In a new study published in Cell , the international research team, led by the Chinese Academy of Agricultural Science and spanning 22 research institutions across 10 countries, has shown that building a pangenome enables breeders to identify and reintroduce beneficial genes and structural DNA variants that were lost when crops were first domesticated.
As a proof of concept, they produced a candidate breeding line of Tartary Buckwheat that includes beneficial alleles from wild species and landraces that both withstand high-altitude stress and produce larger seeds.
In general, genomics-assisted breeding relies heavily on single-nucleotide polymorphisms (SNPs) measured against a single reference genome. That approach is largely blind to structural variation, which can have large effects on traits. Because domestication tended to discard genes useful only in the wildest environments, many beneficial variants survive today only in a crop's wild relatives.
In the study, the team assembled a graph-based pangenome from 16 genomes and catalogued 123,131 structural variants across the species. The pangenome revealed a wild-specific gene that enhances high-altitude adaptability and a specific genetic area that contributes to seed-size variation across wild buckwheat and landraces. Leveraging these superior traits, they developed buckwheat lines with enhanced high-altitude adaptability and improved yields across test sites.
The authors describe the result as a generalisable strategy for integrating beneficial alleles, including copy-number variants, from both wild relatives and local landraces to accelerate elite crop improvement. Given that all major crops have wild relatives and landraces holding variation that single-reference approaches overlook, the researchers argue that the method transfers well beyond the crop they tested.
In a statement, CCFI Director and co-corresponding author of the study, Professor Rajeev Varshney FRS FAA, explained:
"Most modern breeding compares a crop's genome against a single reference, which is like judging a language by a single dictionary. A pangenome captures the whole vocabulary, including the words a crop lost along the way. And that's where a lot of the useful genetics for resilience is hiding."
"Resilience and yield are usually a trade-off, meaning that when you increase one, you lose the other. What the pangenome lets us do is identify the specific DNA segments underlying each trait and deliberately stack them. That's a template other breeding programmes can follow," he said.
The team chose Tartary buckwheat (Fagopyrum tataricum), a hardy, nutrient-dense grain native to the Himalayas and important for food security in high-altitude communities, because its wild relatives survive conditions, including intense UV-B radiation and cold, that most crops cannot.
They then built a gap-free (telomere-to-telomere) reference genome, integrated genetic data from 994 accessions across 15 countries, and assembled a 16-genome pangenome spanning wild Himalayan populations and globally distributed landraces.
The pangenome revealed a gene, FtRNH, present in high-altitude wild plants but absent from all cultivated varieties, that helps plants repair UV-B–induced DNA damage. It also pinpointed a second locus, FtPLATZ, where extra gene copies and a small promoter insertion are linked to larger seeds. By crossing these variants together and using marker-assisted selection, the team produced a stacked candidate line carrying both the wild stress-adaptation gene and the large-seed variants. In high-altitude field trials, it showed better growth and significantly higher yield than the standard variety while also producing larger seeds.
Murdoch University Deputy Vice Chancellor Research and Innovation, Professor Peter Eastwood, said:
"This is a landmark study that demonstrates how data-led breeding strategies can be applied to major staple crop breeding programs to deliver varieties tailored to specific growing conditions. What's most important is the applicability of this approach to crops bred in regions where food-security pressure is greatest, and I look forward to seeing this strategy applied in other CCFI-led projects in Australia and abroad. My congratulations to all the authors for this important piece of research."
Murdoch University Pro-Vice Chancellor, Professor Peter Davies, added:
"CCFI researchers continue to lead the way for genomics-led research that improves grower profitability and food security. But more than that, it's fantastic to see the team go beyond the delivery of genomics resources and establish a replicable, evidence-based breeding strategy that breeders can adopt."
