Table 4 Test for effect of mutations in candidate motifs on gene expression. Table lists the summary statistics for the observed expression level differences of mutated vs. original motif allelic variants for all 16 individual hexamer motifs, the derived five consensus motifs, and the two IMEter motifs (designated by asterisks), and sorted by median values of Cohen’s d. Positive d-values indicated higher gene expression of allelic variants containing the original motif relative to alleles with at least one mutation in it. Associated frequency distributions are shown in Fig. 6. P-values refer to one-sample t-tests to test for greater than zero mean and non-parametric equivalent (Wilcoxon signed-rank test) with applied correction for multiple testing (FDR-false discovery rate). FDR-corrected p-values less than 0.1 (< 10% False Discovery Rate) are highlighted applying bold font. “significant genes” refers to genes with found significant expression differences between the two allelic sets (original motif vs. mutated motif) as detected by two-sample t-tests (not corrected for multiple testing as this was introduced as a filter, not as a strict selection criterion). “number of genes” refers to the number of genes with motifs found present and with two allelic variants (original and mutated). “+/−0.00” indicates zero-rounded positive/negative numbers

From: Identification of cis-regulatory motifs in first introns and the prediction of intron-mediated enhancement of gene expression in Arabidopsis thaliana