BACKGROUND: A promoter-based length polymorphism (5-HTTLPR) of the human serotonin gene (SLC6A4) has exhibited inconsistent association with emotionality phenotypes, such as major depression (MD) and the personality trait neuroticism (N). Several explanations have been posited to account for this discrepancy, including underpowered experimental design and variation in gender ratio, age, and ethnicity. METHODS: Here, we describe three independent tests of association between the 5-HTTLPR locus and both N and MD in samples selected for extremeness of N-score from two homogenous populations (n = 88,142, and 20,921). Calculations of statistical power indicated that at a 5% alpha level, these samples retain 100% power to detect a genetic effect accounting for just .5% of phenotypic variance. Effects of age were regressed out of the phenotypic measure, and gender was included as a covariate. RESULTS: No statistically significant effects of genotype could be identified on either N or MD phenotypes (in all cases, p > or = .26), independently of the genetic mode of action applied. CONCLUSIONS: Our data do not support the hypothesis that the 5-HTTLPR variant contributes significantly toward human emotionality as indexed by either the Eysenck Personality Questionnaire N scale or the DSM-IV for MD.
BACKGROUND: A promoter-based length polymorphism (5-HTTLPR) of the human serotonin gene (SLC6A4) has exhibited inconsistent association with emotionality phenotypes, such as major depression (MD) and the personality trait neuroticism (N). Several explanations have been posited to account for this discrepancy, including underpowered experimental design and variation in gender ratio, age, and ethnicity. METHODS: Here, we describe three independent tests of association between the 5-HTTLPR locus and both N and MD in samples selected for extremeness of N-score from two homogenous populations (n = 88,142, and 20,921). Calculations of statistical power indicated that at a 5% alpha level, these samples retain 100% power to detect a genetic effect accounting for just .5% of phenotypic variance. Effects of age were regressed out of the phenotypic measure, and gender was included as a covariate. RESULTS: No statistically significant effects of genotype could be identified on either N or MD phenotypes (in all cases, p > or = .26), independently of the genetic mode of action applied. CONCLUSIONS: Our data do not support the hypothesis that the 5-HTTLPR variant contributes significantly toward human emotionality as indexed by either the Eysenck Personality Questionnaire N scale or the DSM-IV for MD.
Authors: S Shifman; A Bhomra; S Smiley; N R Wray; M R James; N G Martin; J M Hettema; S S An; M C Neale; E J C G van den Oord; K S Kendler; X Chen; D I Boomsma; C M Middeldorp; J J Hottenga; P E Slagboom; J Flint Journal: Mol Psychiatry Date: 2007-07-31 Impact factor: 15.992
Authors: Marcus R Munafò; Nelson B Freimer; Whitney Ng; Roel Ophoff; Juha Veijola; Jouko Miettunen; Marjo-Riitta Järvelin; Anja Taanila; Jonathan Flint Journal: Am J Med Genet B Neuropsychiatr Genet Date: 2009-03-05 Impact factor: 3.568
Authors: Bryan Maloney; Balmiki Ray; Elizabeth P Hayden; John I Nurnberger; Debomoy K Lahiri Journal: Psychiatr Genet Date: 2009-04 Impact factor: 2.458