I'd like to call nucleosomes from yeast MNase-seq data. In particular, I'm interested in using the PING (Zhang et al., 2012) bioconductor package. The PING documentation discusses all of the post-mapping analysis steps. My question is, however, are there any important pre-processing steps I should be aware of? I'm already filtering out multi-mappers and reads with low quality mapping scores. Typical ChIP-seq peak callers (like MACS) remove all redundant reads (e.g. reads with identical 5' and 3' coordinates). For nucleosome calling, should I also use strict read de-duplication? Or should I use something like FixSeq (Hashimoto et al., 2014) to remove certain redundant reads? Anything else?
Naively, I expect lots of redundant reads, at least those that correspond to well-positioned nucleosomes.
Thanks in advance for any insight.