This Article Abstract Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by McCaffery, J. M. Articles by de Geus, E. Search for Related Content PubMed PubMed Citation Articles by McCaffery, J. M. Articles by de Geus, E.

Genetics in Psychosomatic Medicine: Research Designs and Statistical Approaches

From the Weight Control and Diabetes Research Center (J.M.M.), Brown Medical School and The Miriam Hospital, Providence, RI; Georgia Prevention Institute (H.S., Y.D.) and Department of Pediatrics, Medical College of Georgia, Augusta, GA; Twin Research and Genetic Epidemiology Unit (H.S.), St. Thomas’ Hospital, London, UK; and the Department of Biological Psychology (E.D.G.), Vrije Universiteit, Amsterdam, The Netherlands.

View larger version (15K): [in this window] [in a new window]   Figure 1. Variance decomposition in twin studies using a path model. Following standard path tracing rules, the expectation for the variance and covariances obtains as: Variance (SBP) = (a*1*a) + (c*1*c) + (e*1*e) = a2 + c2+ e2; monozygotic covariance (SBP) = (a*1*a) + (c*1*c) = a2 + c2; dizygotic covariance (SBP) = (a*0.5*a) + (c*1*c) = 0.5a2 + c2. Path coefficients a, c, e can be estimated by a maximum likelihood estimation procedure that optimally fits the expected variance and covariances to the observed variance in SBP and the observed cross-twin SBP covariances in MZ and DZ twin pairs. SBP = systolic blood pressure.

View larger version (15K): [in this window] [in a new window]   Figure 2. Bivariate twin model of the tracking of systolic blood pressure (SBP) across time. From the bivariate path diagram, we can compute the genetic correlation between the two time points (rg) as the genetic covariance divided by the square root of the genetic variances of both traits (a11*a21/(a112*(a212 + a222)). The unique environmental correlation obtains as e11*e21/(e112*(e212 + e222)).

View larger version (16K): [in this window] [in a new window]   Figure 3. Twin model testing gene-environment interaction in the presence of gene-environment correlation. Ac is a hypothetical set of pleiotropic genes that lower SBP and increase the drive to exercise regularly. As are genes specific to SBP. Regular exercise is simultaneously allowed to act as an environmental modulator of the genetic effects on SBP (ß1, ß2) as well as of the unique environmental effects on SBP (ß3, ß4). Significance of the gene-exercise interactions can be tested by comparing this full model to models with ß1 or ß2 set to zero. SBP = systolic blood pressure.

View larger version (25K): [in this window] [in a new window]   Figure 4. Three-SNP haplotypes and genotypes. The upper part of the figure shows the eight haplotypes that can be formed from three SNPs that are in full linkage equilibrium (the general rule is 2n with n = the number of SNPs). These eight haplotypes give rise to the 36 different genotypes (the general rule is ((m + 1)*m)/2 with m the number of haplotypes). Because genotyping does not discriminate between paternal or maternal alleles, the upper part of the matrix (blue) has identical genotypes as the lower part of the matrix. The genotypes printed boldface unambiguously translate to haplotypes. All other genotypes printed in black can derive from multiple combinations of haplotypes. If the SNPs are in linkage disequilibrium, only a few of these possible haplotypes and genotypes will be observed in the population. SNP = single nucleotide polymorphism.