Hi,
I have 4 human ChIP-seq datasets, including drug A treated H3K4me3 and IgG, and drug B treated H3K4me3 and IgG. Meanwhile, sacCer3 spike-in data is added to human ChIP-seq data.
Here is what I did:
After aligning, I got 8 bam files, including drug A treated H3K4me3 hg38 bam and sacCer3 bam, drug A treated IgG hg38 bam and sacCer3 bam, drug B treated H3K4me3 hg38 bam and sacCer3 bam, drug B treated IgG hg38 bam and sacCer3 bam Then, I got 4 bed files, including drug A treated H3K4me3 and IgG bed files and drug B treated H3K4me3 and IgG bed files which are scaling by spike-in bam.
Next, I'd like to find differential peak between drug A treated H3K4me3 and drug B treated H3K4me3. Firstly, I tried to use diffbind r package for this part. However, it seems diffbind required me provide peak bed files. I'd like to use homer to do peak calling. However, homer required me to provide two bam files: for example, one drug A treated H3K4me3 bam and one drug A treated IgG bam. However, I have difficulty generating these two bam files. As you know, I do have drug A treated H3K4me3 hg38 bam and sacCer3 bam, drug A treated IgG hg38 bam and sacCer3 bam, but I don't know how to generate one scaled or normalized drug A treated H3K4me3 bam and one drug A treated IgG bam.
Please advise if possible. Thanks!
what method you have used for normalization of your chip-seq data? Go through this post for the same How To Do Normalization Of Two Chip-Seq Data?