Abstract
BackgroundA multi-population genomic prediction (GP) model in which important pre-selected single nucleotide polymorphisms (SNPs) are differentially weighted (MPMG) has been shown to result in better prediction accuracy than a multi-population, single genomic relationship matrix ({mathbf{GRM}}) GP model (MPSG) in which all SNPs are weighted equally. Our objective was to underpin theoretically the advantages and limits of the MPMG model over the MPSG model, by deriving and validating a deterministic prediction equation for its accuracy.MethodsUsing selection index theory, we derived an equation to predict the accuracy of estimated total genomic values of selection candidates from population A (r_{{{mathbf{EGV}}_{{A_{T} }} }}), when individuals from two populations, A and B, are combined in the training population and two {mathbf{GRM}}, made respectively from pre-selected and remaining SNPs, are fitted simultaneously in MPMG. We used simulations to validate the prediction equation in scenarios that differed in the level of genetic correlation between populations, heritability, and proportion of genetic variance explained by the pre-selected SNPs. Empirical accuracy of the MPMG model in each scenario was calculated and compared to the predicted accuracy from the equation.ResultsIn general, the derived prediction equation resulted in accurate predictions of r_{{{mathbf{EGV}}_{{A_{T} }} }} for the scenarios evaluated. Using the prediction equation, we showed that an important advantage of the MPMG model over the MPSG model is its ability to benefit from the small number of independent chromosome segments (M_{e}) due to the pre-selected SNPs, both within and across populations, whereas for the MPSG model, there is only a single value for M_{e}, calculated based on all SNPs, which is very large. However, this advantage is dependent on the pre-selected SNPs that explain some proportion of the total genetic variance for the trait.ConclusionsWe developed an equation that gives insight into why, and under which conditions the MPMG outperforms the MPSG model for GP. The equation can be used as a deterministic tool to assess the potential benefit of combining information from different populations, e.g., different breeds or lines for GP in livestock or plants, or different groups of people based on their ethnic background for prediction of disease risk scores.
Highlights
A multi-population genomic prediction (GP) model in which important pre-selected single nucleotide polymorphisms (SNPs) are differentially weighted (MPMG) has been shown to result in better prediction accuracy than a multi-population, single genomic relationship matrix ( genomic relationship matrices (GRM) ) GP model (MPSG) in which all SNPs are weighted
The equation can be used to assess the potential benefit of combining information from different populations, e.g., different breeds or lines for GP in livestock or plants, or different groups of people based on their ethnic background for the prediction of disease risk scores
We found that the estimate of genomic heritability from the GREML model [37, 38] was equivalent to ρA21(results not shown). de los Campos et al [39] showed that when all causal mutations are included in the GREML analysis, the genomic heritability parameter is equivalent to the proportion of genetic variance explained by SNPs (ρ 2)
Summary
A multi-population genomic prediction (GP) model in which important pre-selected single nucleotide polymorphisms (SNPs) are differentially weighted (MPMG) has been shown to result in better prediction accuracy than a multi-population, single genomic relationship matrix ( GRM ) GP model (MPSG) in which all SNPs are weighted . E.g., Raymond et al Genet Sel Evol (2020) 52:21 numerically small breeds or lines in livestock or numerically small human ethnic groups, it is difficult or impossible to assemble a large enough training population that can accurately predict the genomic values. A potential option to increase the accuracy of GP in numerically small populations is to use a large training population made up of individuals from multiple populations, including the target population, a method known as multi-population GP. Results from dairy cattle indicate that this approach can lead to substantial increases in the accuracy of GP for numerically small breeds, if the training population is made up of individuals from different but closely-related breeds that have recently had substantial exchanges of genetic material, and that a large number of individuals from the additional breed is included [10]. In cases in which distantly related breeds were combined in a single training population, increases in the accuracy of multi-population GP were limited compared to that of within-population GP [11,12,13]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
