Application of a Bayesian non-linear model hybrid scheme to sequence data for genomic prediction and QTL mapping

Abstract

Background: Using whole genome sequence data might improve genomic prediction accuracy, when compared with high-density SNP arrays, and could lead to identification of casual mutations affecting complex traits. For some traits, the most accurate genomic predictions are achieved with non-linear Bayesian methods. However, as the number of variants and the size of the reference population increase, the computational time required to implement these Bayesian methods (typically with Monte Carlo Markov Chain sampling) becomes unfeasibly long.

Results: Here, we applied a new method, HyB_BR (for Hybrid BayesR), which implements a mixture model of normal distributions and hybridizes an Expectation-Maximization (EM) algorithm followed by Markov Chain Monte Carlo (MCMC) sampling, to genomic prediction in a large dairy cattle population with imputed whole genome sequence data. The imputed whole genome sequence data included 994,019 variant genotypes of 16,214 Holstein and Jersey bulls and cows. Traits included fat yield, milk volume, protein kg, fat% and protein% in milk, as well as fertility and heat tolerance. HyB_BR achieved genomic prediction accuracies as high as the full MCMC implementation of BayesR, both for predicting a validation set of Holstein and Jersey bulls (multi-breed prediction) and a validation set of Australian Red bulls (across-breed prediction). HyB_BR had a ten fold reduction in compute time, compared with the MCMC implementation of BayesR (48 hours versus 594 hours). We also demonstrate that in many cases HyB_BR identified sequence variants with a high posterior probability of affecting the milk production or fertility traits that were similar to those identified in BayesR. For heat tolerance, both HyB_BR and BayesR found variants in or close to promising candidate genes associated with this trait and not detected by previous studies.

Conclusions: The results demonstrate that HyB_BR is a feasible method for simultaneous genomic prediction and QTL mapping with whole genome sequence in large reference populations.

Fig. 1

The pseudo-code of the EM module

Fig. 2

The computational time comparison between GBLUP, BayesR and HyB_BR on 600 K and SEQ data. Three reference sets (Ref1, Ref2 and Ref3) with the same number of variants (600 K or SEQ) are used here. Ref1 has Holstein bulls data with 3049 animals; Ref2 has Holstein bulls and cows data with 12,527 animals; Ref3 has Holstein and Jersey bulls and cows with 16,214 individuals

Fig. 3

The prediction accuracy of GBLUP, BayesR, and HyB_BR on 600 K and SEQ data related to three milk production traits including Fat Yield (a), Milk Yield (b), Protein Yield (c), Fat Percent (d), and Protein Percent (e)

Fig. 4

Posterior possibilities of all the variants on fat yield estimated from BayesR (a) and HyB_BR (b) according to their positions (base pairs) across the whole genome. The top SNPs with highest posterior possibilities are labelled with blue circle

Fig. 5

Posterior possibilities of all the variants for milk yield estimated from BayesR (a) and HyB_BR (b) according to their positions (base pairs) across the whole genome. The top SNPs with highest posterior possibilities are labelled with blue circle

Fig. 6

Posterior possibilities of all the variants for protein yield estimated from BayesR (a) and HyB_BR (b) according to their positions (base pairs) across the whole chromosome genome. The top SNPs with highest posterior possibilities are labelled with blue circle

Fig. 7

Posterior possibilities of all the variants for fat percent estimated from BayesR (a) and HyB_BR (b) according to their positions (base pairs) across the whole genome. The top SNPs with highest posterior possibilities are labelled with blue circle

Fig. 8

Posterior possibilities of all the variants on fertility estimated from BayesR (a) and HyB_BR (b) according to their positions (base pairs) across the whole genome. The top SNPs with highest posterior possibilities are labelled with blue circle

Fig. 9

Mapping posterior probabilities of all the variants estimated from BayesR (a) and HyB_BR (b) according to their positions (base pairs) across the whole chromosome related to Fat yield affected by heat tolerance. The top SNPs with highest posterior possibilities are labelled with blue circle

Fig. 10

Mapping the posterior probabilities of all the variants estimated from BayesR (a) and HyB_BR (b) according to their positions (base pairs) across the whole chromosome related to Milk yield affected by heat tolerance. The top SNPs with highest posterior possibilities are labelled with blue circle

Fig. 11

Mapping the posterior probabilities of all the variants estimated from BayesR (a) and HyB_BR (b) according to their positions (base pairs) across the whole chromosome related to protein yield affected by heat tolerance. The top SNPs with highest posterior possibilities are labelled with blue circle