Accuracy of genomic values predicted using deregressed predicted breeding values as response variables

Abstract

Highly accurate predicted genetic values must be obtained at an early age to promote rapid genetic progress. The objectives of this study were to compare accuracies (R 2 ) of genomic values (GVs) and to estimate genetic correlation between true genetic values and genomic values obtained using predicted breeding values (EBV) and deregressed EBV (DEBV) as response variables. A first population, effective population size 800 and 100 generations, was simulated using the QMSim program to generate linkage disequilibrium. Thereafter, 20 males and 200 females were used to generate a second 14-generation population, with 6,400 individuals per generation and its corresponding phenotype and genotype in SNP terms. Generations 7 to 14 of the second population were used in several combinations as training (PEn) and evaluation (PEv) subpopulations. GVs, their accuracies, and genetic correlations were obtained using the GenSel and ASREML programs. When PEn was the largest, the mean R 2 of GV was the highest, 0.77 ± 0.01. The closer PEn was to PEv, the higher the R 2 , and correspondingly, the lower the predicted error variance. The trends for R 2 and PEV held true for both EBV and DEBV used as response variables. Genetic correlation estimates between true genetic values and GVs varied from 0.41 to 0.53 in the two scenarios studied. They decreased when PEn and PEv were farther apart. There were only slight advantages of using DEBVs as response variables over using EBVs.KEY WORDS: Genomic evaluation, Deregressed predicted genetic value, Genomic predicted value, Accuracy, Genetic correlation. RESUMENLos valores genéticos de individuos en una población deben obtenerse de forma precisa y a edad temprana para promover un progreso genético rápido. Los objetivos de este estudio fueron comparar las exactitudes (R 2 ) de valores genómicos predichos (GBV) y estimar la correlación genética entre los valores genéticos verdaderos (TGV) y los GBV, utilizando los valores genéticos estimados (EBV) y EBV ajustados (DEBV) como variables respuesta. Una primera población de 100 generaciones con tamaño efectivo 800 se simuló con el programa QMSim para generar desequilibrio de ligamiento. Posteriormente, se utilizaron 20 machos y 200 hembras por generación en una segunda población de 14 generaciones, con 6,400 individuos por generación y sus correspondientes fenotipos y genotipos en términos de SNP. Las generaciones 7 a 14 de la segunda población se usaron como subpoblaciones de entrenamiento (PT) y evaluación (PE). Los GBV, sus exactitudes y correlaciones genéticas se obtuvieron utilizando los programas GenSel y ASREML. Cuando la PT fue la más grande, R 2 media fue la más alta, 0.77 ± 0.01. Cuanto más cercana es PT a PE, mayor R 2 , y menor la varianza del error de predicción (PEV). Las tendencias para R 2 y PEV se mantuvieron tanto para EBV como para DEBV utilizadas como variables de respuesta. Los estimadores de correlación genética entre TGV y GBV variaron de 0.41 a 0.53 en los dos escenarios estudiados. La R 2 disminuyó cuando PT y PE es...

GWAS-Based Identification of New Loci for Milk Yield, Fat, and Protein in Holstein Cattle

“…In dairy breeds, EBVs or deregressed EBVs are preferred as the dependent variable of GWAS. Research has shown that the deregressed EBVs could reliable in genomic analysis, but some studies showed that the accuracy of genomic analysis when using EBVs was only slightly lower than using deregressed EBVs [35,36]. In addition, a simulation study indicated that GWAS using EBVs or deregressed EBVs as the dependent variable with polygenic effect modeled had similar performance in controlling the false positive rates (FPR), even when using deregressed EBVs lowered the power in some degree [37].…”

Estrutura populacional e acurácia da predição genômica em populações de aves: um estudo de simulação

“…, permitindo avaliar o grau de similaridade entre as previsões (Momen et al, 2018;Ramírez-Flores et al, 2017). Foram utilizados as médias e os errospadrões das dez repetições do processo de simulação.…”