Genomic-enabled Prediction Accuracies Increased by Modeling Genotype × Environment Interaction in Durum Wheat

Area of Science:

• Plant breeding
• Genetics
• Agricultural science

Background:

• Genomic prediction studies in durum wheat often overlook genotype × environment (G×E) interactions.
• Incorporating G×E effects is crucial for accurate prediction in diverse field conditions.

Purpose of the Study:

• To evaluate the prediction accuracy (PA) of genomic selection models with and without G×E effects in durum wheat.
• To compare different cross-validation (CV) schemes for predicting performance in various environments.

Main Methods:

• Tested six non-G×E and three G×E Genomic Best Linear Unbiased Predictor (GBLUP) models.
• Employed three CV schemes (CV0, CV1, CV2) across yield potential, drought, and heat stress conditions.
• Analyzed eight traits including grain yield, plant height, and NDVI.

Main Results:

• G×E models demonstrated higher PA than non-G×E models.
• CV2 and CV0-Env schemes yielded higher PA (0.58) compared to CV1 (0.35).
• Plant height showed the highest PA (0.68), while grain yield had moderate PA (0.34).

Conclusions:

• Genomic selection models integrating G×E interactions offer significant potential for forward prediction in durum wheat breeding.
• These models can enhance genetic gains by improving the selection of superior lines across environments.