Both serotonin transporter (SLC6A4) and brain-derived neurotrophic factor (BDNF) genes have shown positive associations with obsessive-compulsive disorder (OCD) and some other psychiatric disorders, but these results have not been consistently replicated. To explore the hypothesis that this variability might result from the effects of differing combinations of overlooked variants within SLC6A4 together with small OCD and control sample sizes, we studied three common functional polymorphisms (5-HTTLPR, STin2, and the newly discovered SNP, rs25531) in the largest sample size of OCD patients (N=347) and controls (N=749) ever investigated. During methods development, we found evidence for potential SLC6A4 genotyping problems with earlier methodology, a third possible contributor to variability in earlier studies. A fourth possible explanation might be SLC6A4 x BDNF interactions, which prompted us to investigate combined genotypes of BDNF V66M with the three SLC6A4 loci. Except for a nominal association with rs25531 alone, which did not survive correction for multiple comparisons, we found no evidence for any of these other variants being associated alone or together with OCD, and we therefore also examined clinical OCD subtypes within the sample to evaluate clinical heterogeneity. Subgroups based on the age of OCD onset, gender, familiality, factor analysis-derived symptom dimensions, or comorbidity with other psychiatric disorders failed to identify SLC6A4- or BDNF-associated phenotypes, with one exception of overall number of comorbid anxiety disorders being significantly associated with 5-HTTLPR/rs25531. We conclude that despite their attractiveness as candidate genes in OCD, our data provide no support for association in this large OCD patient sample and point toward the need to examine other genes as candidates for risk determinants in OCD.