Entering edit mode
4.4 years ago
Morris_Chair
▴
370
Hello everyone,
I was watching a tutorial about how to use MACS, one step is for sampling down treatment and control in order to have the same amount of reads. Do you guys think that I should sample down everything? including input, histone acetylation and methylation as well?
thank you
Why do you feel that it might be warranted for your data set? Did you understand the rationale for the downsampling in the tutorial?
Hi Friederike, the rational should be to have the same amount of reads in all samples and avoid extra peaks in samples with more reads. let say that with the protein of interest I have a file of 1 million of reads, input 1.5 and histone acetylation 0.8 million I think to sample down considering the sample with less reads compared to the rest, in this case I bring all sample to 0.8 million, Is it correct?
thank you
Yes. However, MACS is usually doing the library size correction internally, there shouldn't be a need for you to downsample manually AFAIK