Computational protocol: Examining for an association between candidate gene polymorphisms in the metabolic syndrome components on excess weight and adiposity measures in youth: a cross-sectional study

Similar protocols

Protocol publication

[…] A descriptive analysis was performed on the perinatal history and the socioeconomic, health, anthropometric, and genetic variables. The comparison of the variables among the groups according to nutritional status with the Pearson X 2 and Pearson correlation was made; the Student t tests were made for quantitative variables. For the allele and genotype frequencies, the Hardy–Weinberg equilibrium and the associations of the variant phenotypes, PLINK v1.07 [] were used. Logistic regression analysis was performed to look for associations of polymorphisms with nutritional status (normal weight and overweight groups) using the following models: additive (major allele homozygotes vs. heterozygotes vs. minor allele homozygotes), dominant (major allele homozygotes vs. heterozygotes + minor allele homozygotes), and recessive (major allele homozygotes + heterozygotes vs. minor allele homozygotes). The best model was chosen according to the Akaike information criterion. Logistic regression was employed to estimate the odds ratio (OR) and its confidence interval (CI) of 95%. The association of genotype with BMI, WC, and percentage of BF were evaluated using linear regression; variables with non-normal distributions were log-transformed before analysis. For the anthropometric measurements that were associated to more than two associated variants, an estimation of the individual genetic risk score (GRS) was generated, using the risk alleles. We quantified the unweighted genetic risk score to assess the combined effects of the variants by summing the number of risk alleles; each individual might have 0, 1, or 2 risk alleles in each of the variants. The weighted GRS was calculated by multiplying the number of risk alleles at each locus (0, 1, 2) by the variants’ β coefficient from the predictive model and then summing the product. Effects of covariables (gender, age, pubertal maturation, and BMI) were controlled, as well as the interaction between the genetic variants and the social stratum, maternal education year, birth weight, and maternal breastfeeding. In addition to the previous analysis, we conducted logistic analyses for the subset of the samples for which information ancestry was accessible. The objective of these analyses was to determine if the possible association of the polymorphism with nutritional status (normal weight and overweight groups) or adiposity measurement could be driven by population stratification. Controlling for two out of three admixture estimates avoided collinearity in the model, since the three ancestry components sum up to 1.Multiple test correction was done via permutation tests; we performed 10,000 permutations to determine empirical significance, which was considered for a value of p < 0.05. […]

Pipeline specifications

Software tools PLINK, ADMIXTURE
Applications Population genetic analysis, GWAS
Organisms Homo sapiens