search

biocore-ntnu / epic

(DEPRECATED) epic: diffuse domain ChIP-Seq caller based on SICER
http://bioepic.readthedocs.io
MIT License
31 stars 6 forks source link

epic - bed formats #67

Closed sejmodha closed 6 years ago

sejmodha commented 6 years ago

Hi There,

I have used epic for ChIP-Seq data and now I want to annotate peaks using peak annotation program e.g. annotatePeaks (from Homer), could you please explain the epic's default output format and the bed format generated using -b flag as the bed file does not have strand or peak ID info in it.

endrebak commented 6 years ago

I'd love to help you make the output homer-compatible. What do you need the data to look like?

Each enriched region (peak) is a separate line in the bed. ChIP-Seq data is typically not stranded, and the epic data is not either: the counts from each strand are merged into each peak. . is a common way of writing "no info" in bed files.

The bed epic writes is especially tailored for use in the UCSC genome browser.

This describes the bed epic produces: http://bioepic.readthedocs.io/en/latest/output_files.html#b-bed-optional

The example bed is here for some reason: http://bioepic.readthedocs.io/en/latest/output_files.html#i2bw-individual-log2fc-bigwigs (I need to move it to the right section :/ )

sejmodha commented 6 years ago

I see! that makes sense, sorry I didn't visit the readthedocs page.
I can re-format the data to be in the specific format required by annotatePeaks http://homer.ucsd.edu/homer/ngs/annotation.html. Thanks for the prompt response.