Fig. 4
From: MegaLMM: Mega-scale linear mixed models for genomic predictions with thousands of traits
Average within-trial prediction accuracy for four maize traits in the Genomes2Fields Initiative experiment. Traits included: days to silking (DTS), anthesis-silking interval (ASI), grain yield, and plant height. Bars show the average ±95% confidence intervals of prediction accuracy for each method across the 76-99 trials with sufficient training data for each trait. For each trail, prediction accuracies were estimated as the mean over 20 randomized cross-validation replicates