|Abstract:||Since its discovery in 1969, Goss’s wilt, a foliar blight and vascular wilt disease caused by the gram-positive bacterium Clavibacter michiganensis subsp. nebraskensis (Cmn), has emerged as one of the top four diseases of maize in the United States and Canada. No source of complete resistance has been described for Goss’s wilt and little is known about the genetic and mechanistic basis of host resistance to Cmn. The objective of this study was to perform a linkage mapping and genome-wide association study (GWAS) to identify regions of the genome associated with Goss’s wilt resistance. Additionally, we sought to use genomic prediction models to evaluate the use of genomic selection in predicting Goss’s wilt phenotypes in a panel of diverse maize lines. Within the Intermated B73 x Mo17 (IBM) population and three disease resistant introgression lines (DRIL) populations: B73 x Mo17, Mo17 x B73, and NC344 x Oh7B, we were able to both identify novel QTL and confirm previous findings. In a GWAS of the Goodman maize diversity panel, we were unable to identify any variants significantly associated with Goss’s wilt. However, using genomic prediction, we were able to train a model with an accuracy of 0.6971. In addition, when evaluating the accuracy of our prediction model under reduced marker density, it was shown that only 10,000 single nucleotide polymorphisms, or ~20% of our total marker set, was necessary to achieve our control model’s prediction accuracy. This is the first report of genomic prediction for a bacterial disease of maize, and these results highlight the potential of genomic selection for disease resistance in maize.