Crossbreeding is practiced in commercial breeding programs of many plant and animal species to exploit heterosis, breed complementarity, and protect pure line genetic material. Success of crossbreeding schemes depends on identifying and using the right combination of pure lines to produce the desired crossbred offspring. Currently, selection of pure lines is based on results of “field tests”, during which the performance of crossbreds is assessed under commercial settings. Field tests are time-consuming, and constitute a large percent of the costs of commercial crossbreeding programs. The research in this thesis set out to develop models for the accurate prediction of heterosis in White Leghorn crossbreds, using 60K SNP genotypes from their parental pure lines. Based on a dominance model, we showed that models using the squared difference in allele frequency between parental pure lines can predict heterosis with an accuracy of ~0.5, cutting the costs of field tests by about 50%.