How to deal with biological replicates of nucleosome MNase-seq data?

[wisense]

Hi, all
Now I got a problem in handling the biological replicates of nucleosome MNase-seq data. There are two biological paired-end MNase-seq data of mouse MEF cells, one replicate reads length is 51bp while another replicate reads length is 101bp. I want call the merged nucleosome occupancy of the MEF cells, how to use the biological replicates data to call the nucleosome occupancy? PS. I used the software GeneTrack to do the peak calling.

yours,
wisense

[jwfoley]

I have a tool designed for this kind of data, which I plan to announce within the next few weeks. Send me a private message including your platform (64-bit Linux, Mac OS X, something else) if you'd like to try it.

[wisense]

I have a tool designed for this kind of data, which I plan to announce within the next few weeks. Send me a private message including your platform (64-bit Linux, Mac OS X, something else) if you'd like to try it.

I'd like to try.Thank you very much~
My platform is 64-bit Linux

[jwfoley]

Followed up via PM.

Also, you may have a systematic bias in your data due to different "mappability" of 51-mers vs. 101-mers: some regions of the genome that might be difficult to map with 51 nt reads might be mappable with 101, so you'll only see them in the latter sample; or, some artifacts of spurious mapping with 51 nt might disappear when the full 101 nt are used, so you'll only see them in the former sample. The simplest solution is to truncate your 101 nt reads to 51 nt before mapping, though of course this wastes data. Or verify that this truncation does not change the results.

[metheuse]

I have three replicates of MNase-seq of H3 sequencing reads, for two conditions of yeast. The aim is to find nucleosome positioning difference at some specific loci. Did you publish your tool already? Could I use it? Thanks.

[jwfoley]

Yes, the tool is called UniPeak and it is published in doi:10.1186/1471-2164-14-720. I haven't tried it on MNase-seq data but perhaps it will perform similarly to transcription factor ChIP-seq. If not, try tweaking the parameters, or ask me for help.

[metheuse]

Yes, the tool is called UniPeak and it is published in doi:10.1186/1471-2164-14-720. I haven't tried it on MNase-seq data but perhaps it will perform similarly to transcription factor ChIP-seq. If not, try tweaking the parameters, or ask me for help.

Thanks! I'll let you know if I have questions.