Skip to main content
Fig. 6 | Genetics Selection Evolution

Fig. 6

From: Dimensionality of genomic information and its impact on genome-wide associations and variant selection for genomic prediction: a simulation study

Fig. 6

Prediction accuracy (correlation between true and estimated breeding values) for PBLUP or ssGBLUP using training size based on the percentage of variance explained by eigenvalues (EIGx, with x = 50%, 90%, 95%, 98%, and 99%) or all genotyped individuals (ALL) for different Ne, QTN, and heritability. Results in each training set are the average of all genotyped scenarios: 50k, TOP10, TOP50, TOP100, TOP200, TOP400, and ‘SIG’ for Q200 and 50k, TOP10, TOP100, TOP500, TOP1000, TOP2000, TOP4000, and ‘SIG’ for Q2000 scenarios except for PBLUP

Back to article page