For the significant difference found in multiple subgroups, this test does not enable identification of which multiple subgroups are significantly different, only that across all the subgroups there are differences.
Probability values were adjusted for multiple testing using the False Discovery Rate (FDR) method [ 50].This method uses a controlled FDR while adjusting for testing simultaneously across multiple subgroup comparisons of sex and diet groups.
Multiple subgroups were systematically tested and the ranking of genes was compared.
A Bonferroni-adjusted significance level of 0.0083 was calculated to compare SWVs and ARs of the mass of multiple subgroups (6 hypotheses tested).
Moreover, post hoc implementation of multiple subgroup analyses considers a set of statistical inferences simultaneously (multiple testing), and errors, such as incorrectly rejecting the null hypothesis, are likely to occur.
Such tests for heterogeneity were nominally statistically significant, but only when not adjusted for the multiple subgroup comparisons.
The second section briefly describes the main statistical tests used in clinical trials, as well as certain situations that may increase the risk of false positive findings (type 1 error), such as multiple, subgroup, intermediate and non-inferiority analysis.
However, we performed post hoc testing (with adjustments for multiple testing) when comparing multiple subgroups (genotypes) of a given SNP.
The SWVs and ARs of the masses of multiple subgroups were compared using Kruskal-Wallis test, and any two groups were compared with Mann-Whitney U test.
Spearman correlation coefficient was used to assess the association between two continuous variables, Mann-Whitney test was used for two-group comparisons of continuous variables and Kruskal-Wallis test was used to compare expression of continuous variables in multiple subgroups.
