Tissue-specific transcriptional regulation has diverged significantly between human and mouse
Duncan T. Odom, Robin D. Dowell, Elizabeth S. Jacobsen,
William Gordon, Timothy W. Danford, Kenzie D. MacIsaac,
P. Alexander Rolfe, Caitlin M. Conboy, David K. Gifford and Ernest Fraenkel
Lists of peaks used in the paper
Click here to download
lists of the binding events that were considered signficant at the default cutoffs used in the paper.
Explanation of file format
There are two files, one for human and one for mouse.
Each file is a concatenation of the four transcription factors
(FOXA2, HNF1A, HNF4A, HNF6) for each species. Each peak list begins
with a descriptor line. For example:
---Hs FOXA2:liver:tissue vs WCE:liver:tissue;HC:median
indicates that the data which follows is from Human FOXA2.
Each subsequent line consists of:
TSSid \t #peaks [\t location \t score \t confidence]
TSSid is an identifier of the form refseqid:chromosome:start
which is unique to each transcription start site (TSS).
indicates the number of peaks and/or probes associated with this TSS.
This is followed by zero or more peaks (in no particular order), each
represented by three fields indicating the location, "score" and
"confidence" for each peak. The score is the sum of
(posterior*strength) over width of the peak and the confidence is the
maximum posterior value within the peak. These values are the direct
outputs of JBD (see Qi et al. for more details on the peak calling method). Human genomic coordinates refer to genome version hg17 and mouse coordinates refer to genome version mm6.
Click here to return to the main page for the paper.