We also found that a relatively small number of ERα binding peaks are located in a known 5'TSS region (9% for ChIP-seq, 6% for ChIP-PET, 6% for ChIP-chip).
A big portion of ERα binding peaks are located within intra-genic regions (38% for ChIP-seq, 37% for ChIP-PET, 58% for ChIP-chip) as well as gene desert regions (14% for ChIP-seq, 20% for ChIP-PET, 12% for ChIP-chip), 100 kb far away from known 5'TSSs and 3'TSSs (Additional file 2, Figure S1B).
Whereas ChIP-seq binding peaks were centered on pattern 2 DMRs, pattern 3 DMRs exhibited a marked depletion of RF-binding.
The distribution shown in Figure 4C provides a visual demonstration that LRH-1 binding peaks were enriched close to TSS for known genes.
Significant binding peaks were defined as probes containing four or more signals above background in a 500 bp sliding window; the degree of significance depended on the FDR value.
Thus we have included a considerable portion of CDRs in the datasets, because some binding peaks are located in CDRs.
The binding peaks were determined by the peak finding algorithm provided in the TiMAT package.
Secondary binding peaks were observed on proteins of higher molecular mass that were not recognized by the TSPO antiserum.
For DWG, BEAF-32, CTCF and GAF respectively 4.6%, 14.3%, 17.5% and 33.7% of the binding peaks are located within LADs, while 40.6% is expected by chance.
First, SU HW) binding peaks are preferentially located in the vicinity of LAD borders, with the highest frequency occurring just outside LADs, at ∼4 kb from the borders.
