Optimizing Genomic Prediction Based Sparse Testing Designs for Multi-environment Trials in Tetraploid Potato cultivars

Genomic Prediction (GP) based sparse testing provides an opportunity to evaluate more genotypes within a fixed budget in multi-environment trials (METs) to lower phenotyping costs. It assesses untested genotype-environment combinations using varying training set sizes and allocation schemes. This study implements sparse testing designs in tetraploid potato breeding trials employing varying compositions of overlapping and non-overlapping genotypes in different allocation schemes, and utilizing three calibration set sizes (19, 16, and 13). A total of 114 unique genotypes were tested for dry matter yield (YDY), tuber length (TL), and tuber count (TC), using three GP models (E+L, E+L+G, E+L+G+GE). Considering the largest training set size and the complete non-overlapping and zero-overlapping genotypes strategy, no differences were observed between models, yielding predictive ability (PA) values of 0.83, 0.70, and 0.50 for YDY, TL, and TC, respectively. Conversely, as more overlapping genotypes were included in the designs, PA declined across schemes, with the E+L model showing a more rapid drop compared to models that included genomic data. The similar performance of the E+L+G and E+L+G+GE models indicated that the G×E interaction was minimal in the dataset. Additionally, reducing the training set size had an impact on PA as the designs included fewer non-overlapping genotypes, and there was no restoration of PA after adding more overlapping genotypes to the designs. Our findings suggest that evaluating more non-overlapping and a few overlapping genotypes could reduce phenotyping costs by fivefold and increase the testing capacity for tetraploid potato cultivars in METs for maximizing overall genetic gain.