Ancestry-specific haplotypes are those that have been preserved within specific lineages or breeds as a result of selection pressure, whether that pressure is natural or artificial. These haplotypes based on an individual's ancestry tend to be linked to the most advantageous gene variants for the attributes that are being selected for. The pattern of transmission of these haplotypes may have a direct effect on how individuals manifest their traits. Admixture mapping analysis of crossbreeding populations is one method for determining ancestry specific haplotypes. This dissertation explains how we statistically identified haplotypes and mapped them to their ancestral source thereby evaluating their effect on the genomic architecture and performance of crossbreeding populations. Furthermore, we investigate additional aspects such as the effect of the specific reference population representing the parental breed that may enhance the precision of ancestry mapping. Using the diverse pool of ancestral populations relevant to ProCROSS and Grazecross admixed dairy cattle, we compared variation of genomic breed composition estimates based on distinct reference populations to those derived from pedigrees. We observed that genomic and pedigree-based breed estimations differ by a large margin, and that selecting a reference population closely related to the animals who were crossbred is crucial for obtaining a precise estimate of the genetic composition of the breeds involved. The study also provides evidence of ancestry-specific effects based on variation in global breed composition across crossbred cattle’s performance. In order to learn how selection favors certain breed-specific haplotypes localized to specific chromosomal regions, a larger group of crossbred cattle known as KiwiCross® were analyzed. Three overlapping regions identified from three haplotype-based programs that identify regions of the genome under selection were shown to have a substantial effect on performance. By following the genetic trail back to its ancestry, we learn that beneficial mutations can be breed-specific, as seen on BTA 20 where Holstein ancestry is favored in high performing KiwiCross® and Jersey ancestry is more prevalent in low-performing KiwiCross®. A slightly different scenario was observed on BTA 7 where Holstein ancestry was associated with both high- and low- performing animals but differing haplotypes derived from the Holstein differentiated the performance groups. Given our newfound knowledge of identifying localized breed-specific selection, we investigated whether or not different crossbred populations and management techniques had similar ancestry - specific selection for the same production qualities. Our analysis identified eight segments with an ancestral effect on yield on BTA 6 and 16. In-depth analysis of these haplotypes revealed the presence of production-critical genes previously identified in other association studies. These results provide compelling evidence for the necessity of protecting ancestry-specific haplotypes in order to produce crossbred offspring with optimal performance. Ancestry-specific haplotypes in genomic selection algorithms could improve genomic prediction in crossbred dairy cattle, breeding program design, and genetic improvement.