To address both of your questions:
Interval files with the fdr value can be created by checking the option
"Parse xls files into into distinct interval files".
MACS builds it model using all chromosomes, so features of one
chromosome will affect the outcome from another.
For specific details of the MACS algorithm, it is probably best to
contact the tool authors, the original paper, and/or the MACS
documentation (see links on MACS tool form in Galaxy or search w/ google).
In particular, the FAQ hosted by the authors
, near bottom) is a good
place to start re: the FDR value.
We should also point out that the exact FDR calculation may differ
between versions 1.3 (the version on the main Galaxy public server at
) and MACS version 1.4 (latest released by tool
authors). We noticed that the MACS changelog discusses some
changes/fixes to FDR calculations. This would be a good thing to ask the
MACS authors about - they might be able to give you help to understand
what is in 1.3 vs 1.4.
Apologies for the late reply,
On 9/13/11 8:45 PM, Keith Giles wrote:
To add to my post, I reread the macs documentation and it seems that
the fdr is calculated based on the p value of negative peaks,
regardless of their location. That is to say, a strong peak on one
chromosome could be associated with a high fdr bc of an even stronger
negative peak, anywhere in the genome.
So, is the fdr then the percent of negative peaks relative to total
peaks (pos + neg) at or below a certain p value ?
Sent from my iPhone
The Galaxy User list should be used for the discussion of
Galaxy analysis and other features on the public server
. Please keep all replies on the list by
using "reply all" in your mail client. For discussion of
local Galaxy instances and the Galaxy source code, please
use the Galaxy Development list:
To manage your subscriptions to this and other Galaxy lists,
please use the interface at: