# Integrating genomic prediction into crop DUS testing: new approaches in support of reference collection management and distinctness assessment

## TL;DR

This paper introduces new genomic methods to improve crop variety testing for distinctness, using genetic markers and predictions to reduce costs and improve accuracy.

## Contribution

A novel approach using genomic prediction to support DUS testing and refine distinctness decisions is proposed.

## Key findings

- Genomic prediction can effectively predict variety differences in specific traits for DUS testing.
- The proposed method shows promising results in perennial ryegrass and wheat.
- Integrating genetic data with trial results can enhance distinctness decisions for cross-pollinated crops.

## Abstract

A new approach is proposed for the use of the genetic markers to manage DUS trials, targeted at individual phenotypic characteristics using genomic prediction, as well for supporting Distinctness decisions.

High-performing crop varieties underpin food security. Due to the cost of developing varieties, systems have been established to provide breeders with legal protection for their varieties. In many countries, such protection is afforded by the International Union for the Protection of New Varieties of Plants (UPOV) system. New varieties must be phenotypically Distinct from existing varieties using a set of crop-specific characteristics, as well as Uniform and Stable (DUS). For many crops, DUS assessment is costly as candidates must be compared to many existing varieties in field trials, based on numerous DUS characteristics. The use of genetic markers has long been considered as a potential tool for managing costs of such trials, for example, by identifying existing varieties that need not be compared to candidate varieties. Under UPOV guidance, the use of genetic markers must be reflective of phenotypic differences in DUS characteristics. Within this framework, we propose a new approach for using markers based on the application of genomic prediction, which is used to predict variety differences in individual characteristics. The approach is evaluated with perennial ryegrass and wheat, yielding promising results. Additionally, we propose a novel approach in which genomic prediction is used to refine Distinctness decisions after DUS trials have been run by integrating genetic and trial information. Using perennial ryegrass as an example, we demonstrate that this approach, which respects the primacy of phenotype in DUS testing, could be used to support distinctness decisions, especially for cross-pollinated agricultural crops where Distinctness may be harder to achieve.

The online version contains supplementary material available at 10.1007/s00122-026-05198-6.