There are a number of peak callers available for chIP seq data. We are running MACS and SWEMBL and using overlaps but I would like some feedback about your experiences with chIP seq peak callers. I am aware of the ChIP-Seq Challenge but this was far from a complete study. What do you use?
Here is a paper that might be of interest:
We use our self developed method called GeneTrack (also discussed in the comparison above).
QuEST is a good tool to use for Chip-Seq data analysis, I found it very easy to use and it can handle at the same time TF Chip-Seq and Histone Mark Chip-seq It is however a good practice to use more than one tool and then overlap the results, the paper that Istvan pointed to is a good one because it present an overview of the different approaches used for peak estimation, not all the tools use the same algorithm so we always have to expect difference even not so big but it depends on how stringeant are you in calculating your peaks. QuEST also allow the cration of wig and bed files that you can postprocess or visualize at UCSC Genome Browser.
There is also an interesting paper by Pepke et al called "Computation of Chip-Seq and RNA-seq studies" that could be useful for you
Hope that helps
I collected some callers here https://github.com/crazyhottommy/ChIP-seq-analysis#peak-calling
A second vote for MACS 1.4.2. Aside from being open-source and well-documented, an interesting benefit is that the author tries to force you into the MACS user group in order to download the software. This seems annoying, but it's actually a huge help because Tao Liu quickly responds to user group queries and the discussions are very informative to follow.
Here some peak callers are discussed
Bailey T, Krajewski P, Ladunga I, Lefebvre C, Li Q, et al. (2013) Practical Guidelines for the Comprehensive Analysis of ChIP-seq Data. PLoS Comput Biol 9(11): e1003326. doi:10.1371/journal.pcbi.1003326
I've used BayesPeak running in R. It is much easier to install than MACS (failed for me), which require some (strange to me) files.
But I'm not sure how it handle paired-end data and how to include biological replicates, so I can't recommend it. I'm still searching...