BACKGROUND: Asthma is a complex disease with both genetic and environmental causes. Genome-wide association studies of asthma have mostly involved European populations, and replication of positive associations has been inconsistent. OBJECTIVE: We sought to identify asthma-associated genes in a large Latino population with genome-wide association analysis and admixture mapping. METHODS: Latino children with asthma (n = 1893) and healthy control subjects (n = 1881) were recruited from 5 sites in the United States: Puerto Rico, New York, Chicago, Houston, and the San Francisco Bay Area. Subjects were genotyped on an Affymetrix World Array IV chip. We performed genome-wide association and admixture mapping to identify asthma-associated loci. RESULTS: We identified a significant association between ancestry and asthma at 6p21 (lowest P value: rs2523924, P < 5 x 10(-6)). This association replicates in a meta-analysis of the EVE Asthma Consortium (P = .01). Fine mapping of the region in this study and the EVE Asthma Consortium suggests an association between PSORS1C1 and asthma. We confirmed the strong allelic association between SNPs in the 17q21 region and asthma in Latinos (IKZF3, lowest P value: rs90792, odds ratio, 0.67; 95% CI, 0.61-0.75; P = 6 x 10(-13)) and replicated associations in several genes that had previously been associated with asthma in genome-wide association studies. CONCLUSIONS: Admixture mapping and genome-wide association are complementary techniques that provide evidence for multiple asthma-associated loci in Latinos. Admixture mapping identifies a novel locus on 6p21 that replicates in a meta-analysis of several Latino populations, whereas genome-wide association confirms the previously identified locus on 17q21.