Multi-breed genomic prediction using Bayes R with sequence data and dropping variants with a small effect

Table 3 Overview of scenarios

Scenario	Data	Strategy	DropIter	DropProp	Simulation	Real
HD_FULL_D0	HD	FULL	–	0	Y	Y
S_FULL_D0	SEQ	FULL	–	0	Y	N
S_FULL_D0.25	SEQ	FULL	0, 200 or 10,000	0.25	Y	N
S_FULL_D0.50	SEQ	FULL	0, 200 or 10,000	0.50	Y	N
S_FULL_D0.7	SEQ	FULL	0, 200 or 10,000	0.70	Y	N
S_FULL_D0.9	SEQ	FULL	0, 200 or 10,000	0.90	Y	N
S_CHR_D0	SEQ	CHR	0	0	Y	N
S_CHR_D0.7	SEQ	CHR	10,000	0.70	Y	N
S_CHR_D0.9	SEQ	CHR	10,000	0.90	Y	N
S_KEPT_D0.7	SEQ	KEPT	10,000	0.70	Y	N
S_KEPT_D0.9	SEQ	KEPT	10,000	0.90	Y	Y
S + HD_KEPT + HD_D0.7	SEQ + HD	KEPT + HD	10,000	0.70	Y	N
S + HD_KEPT + HD_D0.9	SEQ + HD	KEPT + HD	10,000	0.90	Y	N

HD = HD genotypes used for prediction, S = sequence variants used for prediction, FULL = all variants analysed together, CHR = all variants analysed per chromosome, KEPT = variants selected by CHR reanalysed with all chromosomes together, KEPT + HD = variants selected by CHR and all HD variants reanalysed with all chromosomes together, dropProp = proportion of variants that is dropped after dropIter MCMC iterations, simulation and real indicate whether the scenario was analysed in the simulated and real datasets

ISSN: 1297-9686