We augmented our set of 515 publicly available epigenomic and sequence feature annotations from our previous study31 with 116 personally curated datasets from NCBI, 2,593 ENCODE histone ChIP–seq datasets and 2,121 ENCODE open chromatin DNase-seq datasets33, all publicly available at the accessions provided in Supplementary Table 2. All files were collected in six-column standard bed file format. This augmentation brought the total number of features to 5,345.
