Table 1.

The details of 227 features used in the integrated model

Features	Cell line	Tissue/cell type	Accession	Feature number	Data type
Histone modifications (H3K4me1, H3K4me3, H3K9me3, H3K36me3, H3K27ac, H3K27me3), PolII and CTCF	E14	ES	GSE31039	8, each with 7 subfeatures, summing up to 56 features	ChIP-seq
Histone modifications (H3K4me1, H3K4me3, H3K9me3, H3K36me3, H3K27ac, H3K27me3), PolII and CTCF		E14.5 whole brain	GSE31039	8, each with 7 subfeatures, summing up to 56 features	ChIP-seq
Histone modifications (H3K4me1, H3K4me3, H3K27ac, H3K27me3), PolII and CTCF		8 week CB	GSE31039	6, each with 7 subfeatures, summing up to 42 features	ChIP-seq
Repeat elements and CpG islands			UCSC mm9	9, each with 7 subfeatures, summing up to 63 features	Bed format
PhastCons most conserved regions, ORF and nucleotide sequences-based features			UCSC mm9	1 conservation feature, 2 ORF features, 7 nucleotide sequences-based features, summing up to 10 features	BED format

Columns represent the feature names, the cell lines of mouse ES cell used (cell Line), the tissue/cell type of the data (tissue/cell type), the NCBI GEO accession number (Accession), feature number and the data type.