Chromatin Immunoprecipitation and Library Synthesis

Chromatin Immunoprecipitation and library synthesis were performed as described [45 (link)]. Briefly, 10⁷ cells were cross-linked with DMEM media with 1% formaldehyde for 15 min and quenched with 2.5 M glycine. After lysis, nuclei were sonicated with a Bioruptor probe (Diagenode) in an ice water bath set to medium strength in increments of 10 s on and 10 s off for 10 min. For the immunoprecipitation (IP), 5 μg of H3K27ac (Abcam ab4729) or 10 μg of H3K4me3 (Abcam ab8580) antibody was used for each replicate. Library synthesis was performed using the ThruPLEX kit (Takara) and quantified using the NEBNext Library Quant Kit (NEB). Sequencing was performed on an Illumina HiSeq2000 sequencer generating an average of 16 million reads per library. Two biological replicates were generated for each cell line. Reads were aligned and processed as described above. Reads for each IP target were merged and a consensus peak set was called using MACS2 with option –broad. For analyses requiring gene transcription start sites (TSS), TSS data in hg19 coordinates were downloaded from BioMart using the R package biomaRt [46 (link), 47 (link)]. For all analyses, only TSS of expressed genes were used. For upset plots comparing the three cell lines, peaks were called within each cell line. Peaks overlapping the respective consensus peak set were then used for plotting.