A new genomic prediction method with additive-dominance effects in the least-squares framework

In our previous work, we proposed a genomic prediction method combing identical-by-state-based Haseman-Elston regression and best linear prediction with additive variance component only (HEBLP|A herein), the most essential component of genetic variation. Since the dominance effects contribute signif...

Full description

Saved in:
Bibliographic Details
Published in:Heredity 2018-08, Vol.121 (2), p.196-204
Main Authors: Liu, Hailan, Chen, Guo-Bo
Format: Article
Language:English
Subjects:
Arabidopsis - genetics
Autosomal dominant inheritance
Computer Simulation
Data processing
Dominance
Genes, Dominant
Genetic diversity
Genetic Markers
Genomics
Genomics - methods
Genotype & phenotype
Heterosis
Least-Squares Analysis
Linear prediction
Models, Genetic
Quantitative Trait Loci
Regression analysis
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In our previous work, we proposed a genomic prediction method combing identical-by-state-based Haseman-Elston regression and best linear prediction with additive variance component only (HEBLP|A herein), the most essential component of genetic variation. Since the dominance effects contribute significantly in heterosis, it is desirable to incorporate the HEBLP with dominance variance component that is expected to enhance the predictive accuracy as we move to the further development: HEBLP|AD, a paralleled implementation of genomic prediction compared with genomic best linear unbiased prediction (GBLUP). The simulation results indicated that when the dominance effects contributed to a large proportion of genetic variation, HEBLP|AD and GBLUP|AD, having similar accuracy, both outperformed HEBLP|A; but when the dominance variation was none or little, HEBLP|A, HEBLP|AD, and GBLUP|AD had similar predictability. The analysis of real data from Arabidopsis thaliana F2 population also demonstrated the latter situation. In summary, HEBLP|AD performed stable whether a trait was controlled by dominance effects or not.
ISSN:0018-067X
1365-2540
DOI:10.1038/s41437-018-0099-5