Clarify Read Inputs and Differential Peak Calling

[quantumdot]

I hope that you can clarify the relationship of reads when supplying multiple treatment and control files to epic. For example, are multiple treatment files pooled (like MACS) or are they treated independently?

Also, I have interest in differential peak calling between treated and non-treated ChIP samples with corresponding input controls. Does epic have this ability like SICER-df.sh, or is there a mechanism to reproduce this functionality?

Thanks for the great work, I am enjoying using epic so far.

[endrebak]

No such df-feature unfortunately. Should not be too hard to implement though.

Input files are pooled after removing duplicate reads. The same is true for ChIP btw. If you want to avoid pooling, you should just run epic multiple times, once on each dataset.

I am finishing my PhD right now, so I do not have time to work much on epic. But it isn't abandoned, will do bug fixes and such.