Effect of non-random mating on genomic and BLUP selection schemes

We studied the effect of mating designs, random, minimum coancestry and minimum covariance of ancestral contributions on rate of inbreeding and genetic gain for schemes with different information sources, i.e. sib test or own performance records, different genetic evaluation methods, i.e. BLUP or genomic selection, and different family structures, i.e. factorial or pair-wise.Results showed that substantial differences in rates of inbreeding due to mating design were present under schemes with a pair-wise family structure, for which minimum coancestry turned out to be more effective to generate lower rates of inbreeding. Specifically, substantial reductions in rates of inbreeding were observed in schemes using sib test records and BLUP evaluation. However, with a factorial family structure, differences in rates of inbreeding due mating designs were minor. Moreover, non-random mating had only a small effect in breeding schemes that used genomic evaluation, regardless of the information source.It was concluded that minimum coancestry remains an efficient mating design when BLUP is used for genetic evaluation or when the size of the population is small, whereas the effect of non-random mating is smaller in schemes using genomic evaluation.Selection and parent mating patterns are the two major components of a breeding program and must be optimised with respect to genetic gain (ΔG) and rate of inbreeding (ΔF). In the literature, several selection and mating designs have been reported that aim at reaching a high ΔG and/or low ΔF [1,2], since a high ΔF represents a risk for the long-term success of breeding programs. While there is a consensus on selection procedures [3,4], many mating designs have been developed, mostly aimed at avoiding the mating of parents that are more related than average [1,5-7].Factorial mating consists in producing half sibs from male and female parents at the expense of full sibs, while preserving offspring numbers per parent [8]; with minimum coa