The following table provide the complete list of candidate enhancer-target
pairs (ETG) inferred in
Salviato et al.
The schematic workflow of our methodological framework is available at
of the paper.
The source code of the method is available in the
Each line corresponds to an ETG pair, where:
- chr, start, end: enhancer region (hg19). Promoter proximal elements are not considered (3.5 kb upstream and 1.5 kb downstream of coding and non-coding TSS).
- symbol: target-gene symbol of protein coding genes, based on RefSeq annotations in UCSC. Target genes with overlapping promoter regions are merged as a single pairs (separated by semicolon).
- distance: distance between mid-points of gene promoter and enhancer regions. Promoters are defined as 1.5kb upstream and 0.5kb downstream regions of TSS of coding genes.
- cca: first canonical correlation coefficient. It is based on the enrichment of DNase-seq and H3K27ac for enhancers and DNase-seq, H3K27ac and H3K4me3 for promoters.
- HC: Hierarchical Contact (HC) score. HC score is proportional to the likelihood of enhancer-promoter pairs co-localization within hierarchy of TADs across multiple Hi-C matrices.
- Pval.raw: -log10 p-value obtained by testing the canonical correlation coefficient. It quantifies the amount of evidence provided by the data for the presence of synchronized activity between the enhancer and promoter gene.
- Pval.adapt: -log10 adjusted p-values by considering the 3D co-localization information encoded in the HC score.
- GTEx, PancanQTL, pcHiC: 1 if supported by the Genotype-Tissue Expression project, the pan-cancer eQTL analysis, or capture Hi-C experiments, respectively.
For more details, please refer to the
Materials and methods
section of the paper.