I have a problem for peak calling. Could anyone help me?
After bowtie, I am using MACS to call peaks. There are 12 million unique reads in input, but there are only 4 million unique reads in my IP. The number of peaks called by MACS is not as high as I expected based on ChIP-chip. I am guessing some of the peaks may be masked because of the more reads in input. How could I solve this problem? Any suggestion?
I plan to randomly select comparable number of reads from input and call peak again. Does it make sense to do so? If it does, how could I do the job? I do not know how to do it.
After bowtie, I am using MACS to call peaks. There are 12 million unique reads in input, but there are only 4 million unique reads in my IP. The number of peaks called by MACS is not as high as I expected based on ChIP-chip. I am guessing some of the peaks may be masked because of the more reads in input. How could I solve this problem? Any suggestion?
I plan to randomly select comparable number of reads from input and call peak again. Does it make sense to do so? If it does, how could I do the job? I do not know how to do it.