Identifying novel Japanese heart failure variants via endothelial cis-regulatory element analysis

Figures

• 

  Figure 1. Overview of the proposed methods.

• 

  Figure 2. Manhattan and Q–Q plots of the Japanese heart failure GWAS. (A) The Manhattan plot displays the negative logarithm of the P-value on the vertical axis, whereas the horizontal axis presents the chromosome number and genomic position. The blue horizontal line denotes P = 1.0 × 10−5, and the red horizontal line signifies P = 1.0 × 10−8. (B) The Q–Q plot depicts the distribution of P-values in the GWAS. The horizontal axis presents the negative log of the expected P-value, and the vertical axis presents the negative logarithm of the observed P-value. The red line reflects the null hypothesis of no true associations.

• 

  Figure 3. Extraction of candidate SNPs located in the CRE of endothelial cells. Functional annotation of SNPs located in the CRE. (A) The diagram presents the numbers of variants found in the promoter and enhancer and the transcription factor-binding regions. (B) The map of candidate SNPs with low P-values on chromosomes among those located in the CRE. Blue pins denote SNPs in the promoter, whereas red pins indicate SNPs in the enhancer.

• 

  Figure 4. Associations of candidate SNPs with histone modification in the genome. (A) The Manhattan plot in proximity to rs3176334 includes visualizations of the LD coefficient, epigenetic modifications (H3K4me3 and H3K27ac) in nine endothelial cell types, and transcription factor-binding regions. The color of the dots in the upper Manhattan plot signifies the strength of the LD coefficient. Enhancer and promoter regions in the endothelial cells are denoted by green and purple bars, respectively, and the ChIP-seq rows represent the transcription factor-binding regions in the endothelial cells. (B) The Manhattan plot near rs12437763 displays the LD coefficient, epigenetic modifications (H3K4me3 and H3K27ac) in endothelial cells, and transcription factor-binding regions. ChIP-seq rows indicate transcription factor-binding regions in both endothelial cells and the heart.

• 

  Figure 5. Transcription factor-binding motifs located near the candidate SNPs. The diagram displays the genomic coordinates of nearby SNPs within the CRE predicted to influence transcription factor motif binding, namely rs3176334, rs78489660, rs35620463, rs2276224, rs1531732, and rs7917309. The positions of these SNPs in the motifs are delineated by dotted boxes in the motif logos and red boxes in the genome sequences, with reference and alternative alleles denoted at the bottom. The genomic sequence and coordinates are presented at the bottom of each display, and the aqua boxes indicate the positions at which the motifs match. The locations of rs3176334, rs78489660, rs35620463, and rs2276224 in the promoters are highlighted in light blue, whereas the locations of rs1531732 and rs7917309 in the enhancers are highlighted in pink.

• 

  Figure 6. Prediction of genes regulated by enhancers in which rs12437763 is located. The upper panel displays the positions of the reference genes. Genes predicted to be regulated by rs12437763 are denoted in purple, other nearby genes are presented in green, and the regulatory genes are represented by gray bars. The bottom panel illustrates chromatin loops predicted to interact with the enhancer, as depicted by purple lines.