Skip to main content
Fig. 3 | Genetics Selection Evolution

Fig. 3

From: Bayesian neural networks with variable selection for prediction of genotypic values

Fig. 3

Sparsity The accuracy of NetSparse versus other methods in scenarios with 10, 100, 300 and 1000 QTL. Each row corresponds to a different amount of sparsity, the different columns correspond to different methods, GBLUP and Bayes-B are additive methods, GBLUP-AD and BayesB-AD are methods with additive and dominance features and GBLUP-ADAA has additive, dominance and additive\(\times\)additive features. The line \(x=y\) is added in red for reference. A marker that is above the line means a replicate with higher accuracy for NetSparse than the method it is compared to, and a marker that is below the line means a replicate with lower accuracy for NetSparse than the other method

Back to article page