Entering edit mode
7.3 years ago
Lila M
★
1.2k
Hi everybody, I'm training in ChIP-seq and I have some doubts that I've could not resolve yet. For example, I'm trying to normalized the read depth profiles of my samples as most advanced papers does. But when I read the extended procedures, I can't find the package or tool that they've used to address this issue. I found some tools, as normalized.loess or deeptools but I can't figure out how the have to be applied. Does any body could give me some advice or tips for that?
Thank you
Many tools handle sequencing depth normalization internally. In general, some sort of normalization should be applied whenever you're comparing samples against each other.
Thank you for your response, I know that, but I would like to improve my knowledge on this field in order to know what kind of tool I have to used. Please, could you be a bit specific? The tool's name, when I have to use them.. Thank you
This is so generic that being specific is impossible. The only tool I know of that doesn't already handle this internally or specify a means of doing this is IGV (and similar programs). There, the signal can be visually scaled, but of course the values aren't. In cases like that use deepTools to produce normalized bigWig files. Everything else that I normally use either automatically handles this (e.g., MACS2 and other peak callers) or has steps defined in its usage where this is handled.
Thank you very much, for example, in MACS2 do you use --SPMR? And after that, can you export it as a bed file to do further analysis as peak annotation?
Thank you
I've never needed the pileup from MACS2.
So, how you do that? Could you explain, please?
I only care about the peak locations and scores.
so you did not make a normalization of read depth profiles, right? you only run MACS2
I want to know how to normalized the read depth profiles equivalent to a number of reads
MACS2 does the depth normalization internally. If you want normalized tracks then just use deepTools (full disclosure, I'm one of the authors).