There have been many conflicting reports concerning the association of the DRD2 locus with alcohol dependence (AD). To investigate whether these findings could be reconciled by considering the genomic region of DRD2 in greater detail, we conducted two separate association studies of AD in 1220 European-American subjects using family-based (488 subjects) and case-control (318 cases and 414 controls) designs, and 43 single nucleotide polymorphisms mapped to the gene cluster of NCAM1, TTC12, ANKK1 and DRD2. We used a generalized linear model and haplotype score tests for the case-control sample, and the family-based association test for the family sample. Haplotype associations centered on TTC12 exon 3 [rs1893699-rs723077; optimal individual haplotype simulated P-value (P(oihs)) = 0.00021] in both independent samples (family and case-control). Additional AD-associated haplotypes centered around NCAM1 exon 12 in the family sample (P(oihs) = 0.0032), and at exons 2 and 5 of ANKK1 in the case-control sample (P(oihs) = 0.00058). LD contrasts between cases and controls support selection at TTC12 exon 3 and ANKK1 exon 2. The armadillo repeat domains encoded by TTC12 and dopamine interact in the Wnt pathway and may have effects on dopamine cell development in the ventral midbrain. We conclude that risk for AD is attributable in part to variants in four regions within this cluster: exon 3 of TTC12, exon 12/intron13 of NCAM1 and exons 2 and 5 of ANKK1. The complexity of these relationships, many of which replicate between our independent samples, may explain prior inconsistent results.