a) Motif enrichment in serum-specific and EpiLSC-specific E1 and E2 distal regulatory elements. Top enriched motifs are shown in the graph. b) Correlation between biological replicates employed in ChIP-seq analysis for ESRRB, NR5A2 and KLF4 transcription factors. Two independent clones were used for each transcription factor and ChIP-seq experiments were performed in serum or 2iDay3 ESCs. Numbers represent the Pearson correlation coefficient. n=18,135 NR5A2-GFP peaks, n=3,743 ESRRB-GFP peaks, n=13,921 KLF4-GFP peaks. c) Motif analysis on ChIP-seq binding sites of ESRRB, NR5A2 and KLF4 transcription factors. High enrichments for known motifs and de novo motifs were found in the corresponding binding sites confirming the specificity of ChIP-seq signals. n=18,135 NR5A2-GFP peaks, n=3,743 ESRRB-GFP peaks, n=13,921 KLF4-GFP peaks. P-values were derived from hypergeometric tests (two-sided). d) Co-occurrence of different transcription factor motifs in E1 and E2 enhancers. e) Average profiles of NANOG, OCT4 and SOX2 for different types of 2i-specific enhancers. As a control, 2,000 random ATAC-seq peaks were used. f) Genomic distribution of differential binding sites (defined in Galonska et al.) of NANOG, OCT4 and SOX2 in serum- and 2i-ESCs. The genomic distribution of these transcription factors was linked to E1-E3 distal regulatory elements identified in this study.