GVCBLUP: a computer package for genomic prediction and variance component estimation of additive and dominance effects

Exploration of genetic variance has mostly been limited to additive effects estimated using pedigree data and non-additive effects have been ignored. This study aimed to evaluate the performance of single nucleotide polymorphisms (SNPs) marker models in the mixed and orthogonal framework including both additive and non-additive effects for estimating variances and genomic prediction in four diabetes-related traits in heterogeneous stock mice. Models have performed differently in detecting SNPs affecting traits. Dominance variances explained over 14.7 and 3.8% of genetic and phenotype variance in a Genomic prediction and variance component estimation method (GVCBLUP) framework. Reliabilities of additive Genomic best linear unbiased prediction model (GBLUP) in different traits ranged from 44.8 to 66.6%, for GVCBLUPs framework including both additive and dominance effects (MAD), and 46.1 to 69% for the model including additive effect (MA). Dominance GBLUP reliabilities ranged from 6 to 26.4% for MAD and from 22.5 to 50.5% in the model including dominance (MD). MA and MD had higher reliability for additive and dominance GBLUPs compared to MAD. Reliabilities of GBLUPs in MAD and MA for all traits were not significant except for growth slope (P < 0.01). In orthogonal framework models, epistasis variances accounted for a greater proportion (87.3, 89.1, 95.5, and 77.2%) of genetic variation for end weight, growth slope, body mass index, and body length, respectively. Heritability in a broad sense was estimated at 1.12, 1.67, 3.64, and 2.0%, in which non-additive heritability had a significant contribution. Genetic variances explained by dominance using GVCBLUPs were 16.8, 29.4, 14.6, and 14.9% for the traits. Generally, the non-additive models had a lower value of deviance information criterion (DIC) and performed better in estimating the variance component. Comparing the estimated variance by orthogonal framework models confirmed the results previously estimated by GVCBLUPs, with the difference that the estimates were shrinking. Following significant SNPs affecting diabetes-related traits by post-genome-wide studies could reveal unknown aspects and contribute to genetic control of the disease.

Download Full-text

Multi-allelic haplotype model based on genetic partition for genomic prediction and variance component estimation using SNP markers

BMC Genetics ◽

10.1186/s12863-015-0301-1 ◽

2015 ◽

Vol 16 (1) ◽

Cited By ~ 11

Author(s):

Yang Da

Keyword(s):

Genomic Prediction ◽

Variance Component ◽

Snp Markers ◽

Variance Component Estimation ◽

Model Based ◽

Component Estimation

Download Full-text

GPS-BDS-Galileo double-differenced stochastic model refinement based on least-squares variance component estimation

Journal of Navigation ◽

10.1017/s0373463321000564 ◽

2021 ◽

pp. 1-16

Author(s):

Hong Hu ◽

Xuefeng Xie ◽

Jingxiang Gao ◽

Shuanggen Jin ◽

Peng Jiang

Keyword(s):

Stochastic Model ◽

Least Squares ◽

Stochastic Models ◽

Variance Component ◽

Satellite System ◽

Variance Component Estimation ◽

Short Baseline ◽

Component Estimation ◽

Zero Baseline ◽

Global Navigation Satellite

Abstract Stochastic models are essential for precise navigation and positioning of the global navigation satellite system (GNSS). A stochastic model can influence the resolution of ambiguity, which is a key step in GNSS positioning. Most of the existing multi-GNSS stochastic models are based on the GPS empirical model, while differences in the precision of observations among different systems are not considered. In this paper, three refined stochastic models, namely the variance components between systems (RSM1), the variances of different types of observations (RSM2) and the variances of observations for each satellite (RSM3) are proposed based on the least-squares variance component estimation (LS-VCE). Zero-baseline and short-baseline GNSS experimental data were used to verify the proposed three refined stochastic models. The results show that, compared with the traditional elevation-dependent model (EDM), though the proposed models do not significantly improve the ambiguity resolution success rate, the positioning precision of the three proposed models has been improved. RSM3, which is more realistic for the data itself, performs the best, and the precision at elevation mask angles 20°, 30°, 40°, 50° can be improved by 4⋅6%, 7⋅6%, 13⋅2%, 73⋅0% for L1-B1-E1 and 1⋅1%, 4⋅8%, 16⋅3%, 64⋅5% for L2-B2-E5a, respectively.

Download Full-text