Figure 3
From: A joint learning approach for genomic prediction in polyploid grasses
Correlation plots for the predicted and real values in the populations: (Pop1) sugarcane (stalk diameter (SD) and stalk height (SH)); (Pop2) Urochloa decumbens (field green weight (FGW), leaf dry matter (LDM), leaf percentage (LP), leaf stem ratio (LSR), regrowth capacity (RC), and total dry matter (TDM)); and (Pop3) Megathyrsus maximus (green matter (GM), LDM, percentage of leaf blade (PLB), RC, stem dry matter (SDM), and TDM). The approaches evaluated were Bayesian ridge regression (BRR), colored green, and the single-environment main genotypic effect model with a Gaussian kernel (SM-GK) with the inclusion of the established feature selection machinery, colored black.
