Evaluation of Models for Utilization in Genomic Prediction of Agronomic Traits in the Louisiana Sugarcane Breeding Program

Sugarcane (Saccharum spp.) is an important perennial grass crop for both sugar and biofuel industries. The Louisiana sugarcane breeding program focused on improving yield by incrementally increasing genetic gain. With the advancement in genotyping (highthroughput) phenotyping techniques, genomic selection a promising marker-assisted tool. In this study, we assessed ridge regression best linear unbiased prediction (rrBLUP) various Bayesian models to evaluate accuracy using 10-fold cross validation 95 commercial elite parental clones from program. Datasets (individual pooled combinations) were constructed based soil type (light—Commerce silty loam, heavy—Sharkey clay) (plant cane, ratoon). A total of 3906 SNPs used predict estimated values (GEBVs) sucrose content cane yield. Prediction was Spearman’s rank correlation Pearson’s between phenotypic GEBVs. All traits showed significant variation with moderate (42% content) high (85% yield) heritability. (0.47–0.80 content; 0.61–0.69 yield, 0.56–0.72 all cross-effect where types considered as fixed effects. general, demonstrated higher than rrBLUP. without effects lower no clear pattern among models. results demonstrate potential implementation variety development