Fig. 6: Replication of the core gene candidate AF association and NKX2-5 target coexpression in independent datasets.

Published proteomics data (PXD006675) as well as RNA-seq data (GSE128188, GTEx) generated from human atrial tissue samples were used for replication. a Centered and scaled values of the mean mRNA or protein expression in AF ctrls and cases, with stronger effects on protein level. GSEA p-values quantify the negative association of NKX2-5 targets with respect to AF. Sample sizes per column: 69 controls, 14 prevalent AF cases, 69 controls, 14 prevalent AF cases (AFHRI, all right atrial appendage); five controls, five AF cases (GSE128188, both right atrial appendage); five controls, five AF cases (GSE128188, both left atrial appendage); three controls, three AF cases (PXD006675, both left atrium). A quantitative description of the qualitative results presented in the heatmap can be found in Supplementary Table 13-14 and Table 3. b Coexpression of NKX2-5 with the 13 identified NKX2-5 transcription factor targets (Pearson’s correlation). Quantified is the correlation between NKX2-5 and its targets on mRNA level for mRNA datasets and the correlation between the NKX2-5 transcript expression with the target protein concentrations for the AFHRI proteomics (NKX2-5 not quantified in proteomics). Sample sizes used for the computation of correlations: 102 AFHRI mRNA, 96 AFHRI protein, 372 GTEx, 10 GSE128188 right, and 10 left atrial appendage samples. AF atrial fibrillation, Ctrl control i.e. individuals in sinus rhythm, GSEA gene set enrichment analysis *Mutation known to affect cardiovascular phenotypes, **Mutation known to affect arrhythmias, ⁺Differential expression or functional impairment for cardiovascular phenotypes, ⁺⁺Differential expression or functional impairment for arrhythmias. Source data are provided as a Source Data file.