Robert Thurman et. al,
"The accessible chromatin
landscape of the human
genome"
Nature, 489,
pp. 75-82, 5 Sep. 2012
Downloadable Data
Much of the data presented in the paper is available via the public
FTP server using anonymous login, at ftp.ebi.ac.uk. See the
supplement for details. This site contains items not available on the
public server.
Further annotation of master list
The paper references
extensively a so-called "master list" of 2,890,742 unique,
non-overlapping DHS positions that span the DHS landscape of all 125
cell-types probed. As described in the supplement, the full master
list (in addition to the DHS called on each individual cell-type) is
available on the EBI public ftp site, at
ftp://ftp.ebi.ac.uk/pub/databases/ensembl/encode/integration_data_jan2011/byDataType/openchrom/jan2011/combined_peaks/
See file
multi-tissue.master.ntypes.simple.hg19.bed
The master list is annotated in column 4 with the number of cell-types
having a DHS overlap at that position. In addition, we have the names
of the cell-types that overlap at each position, in the following
file. There is one row per master list peak, in the same order as the
master list, with cell-type names separated by a semi-colon.
Addendum to Supplementary Table 7
This table of correlated distal/promoter DHS pairs
was created using TSSs of transcripts from Gencode
v. 7.
Only TSSs with confidence levels of 1 and 2 were used (1 = validated, 2 = manually annotated).
Specifically, TSSs with confidence level 3 (automated annotation) were not used. The PAH
gene, shown in Figure 5a of our paper, was labeled "level 3" in Gencode v.7,
and thus does not appear in Supplementary Table 7. The distal DHSs correlated with the PAH
promoter DHS can be found here.
Promoter Predictions
The public FTP site appears to have some inaccessible subdirectories
with regards to the DNaseI/H3K4me3 promoter predictions. Therefore, we
include the full set of predictions here.