This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Stuck with my analysis (MNase-seq)

I am relatively new to NGS analysis and I feel as though I am getting in over my head. I have watched several tutorials, but I can't seem to find information/examples of the kind of analysis that I am interested in doing. I am hoping that someone here might be able to help me out. Our bioinformatics core is very good, but I would like to learn this analysis myself because it will definitely be a good skill to have in the future.

Our lab has recently used MNase-seq with the hopes of looking at changes that occur in chromatin organization under a variety of conditions. What we would like to do is compare changes that occur in our control vs. treated samples. We are interested in global changes such as gain/loss of peaks as well as changes in peak intensity and peak width. I would also like to cross reference these changes with a published ChIP-seq track of our favorite transcription factor to see if these changes occur more preferentially in the vicinity of these transcription factor binding sites.

So far I have been using Galaxy for my analysis. I have the data files from the bioinformatics core in .bed, .bam, .bai, and .fastq formats. I have also looked at the bam files in IGV and have done some manipulations (intersect/window/subtraction) using the bedtools options, but I haven't been able to produce an actual file of these manipulations to do any sort of statistical analysis/comparison. In Galaxy I haven't been able to do much with any these files since Galaxy doesn't seem to recognize the formats despite trying several conversions. I appreciate any input or suggestions that you can offer. In the meantime I will continue plugging away with my data and see if I can figure this out.

mnase-seq mnase

2 answers

Have a look at diffreps and MACS

http://crazyhottommy.blogspot.com/2014/05/macs-for-dnase-seq-data.html

Thanks for the help.

Take a look at danpos. And maybe ngs.plot. Both are pretty easy to run and could get you a bit farther in your investigation.

Thanks for the input. I will look into this.

Thank you for the suggestion. I think that danpos will do exactly what I need, but now I have to learn how to do things for the command line which has been a problem for me in the past.

Log in to answer this question.