Marker-assisted selection (MAS) has been shown, in theory, to produce greater selection gains than phenotypic selection for normally distributed quantitative traits. Theory is presented in this paper for estimating the probability of selecting one or more superior genotypes by MAS (Pr-MAS). This paramater was used to estimate the cost efficiency of MAS relative to phenotypic selection (E-c). Pr-MAS and E-c are functions of heritability (h(2)), heritability of a MAS index (h(I)(2)), the phenotypic selection threshold (i), the genotypic superiority threshold (g), and p = sigma(M)(2)/sigma(G)(2), where sigma(M)(2) is additive genetic variance associated with markers and sigma(G)(2) is additive genetic variance, h(I)(2) increases as p increases. Heritability can be increased to 1.0 by increasing p to 1.0; however, estimated marker effects ((p) over cap) and true quantitative trait locus effects (p) must be perfectly correlated to achieve this in practice. Pr-MAS increases throughout the range of p when i greater than or equal to g, decreases as g increases, and increases as i increases for most p. The frequency of superior genotypes among selected progeny increases as selection intensity increases. E-c ranged from 1.0 to 16.7 for i and g from 1.282 to 2.326, h(2) from 0.1 to 1.0, and p from 0.0 to 1.0; thus, a breeder using phenotypic selection must test 1.0 to 16.7 times more progeny than a breeder using MAS to be assured of selecting one or more superior genotypes. E-c increases as g or i increase and h(2) decreases and increases asp increases when i = g. E-c predicts that MAS substantially decreases the resources needed to accomplish a selection goal for a low to moderate heritability trait when the selection goal and the selection intensity are high.
