how to normalize the bam file for TSS colocalization plotting

[jianhuang lin]

Hi Tao and everyone,

I know that in SeqMINER only the clustering is interfered by normalization, if I have two bam files with different reads numbers. I would like to plot the colocalization of the reads near TSS sites. If the reads numbers are different, how can we compare the peaks near TSS? Is there anyone knows how to normalize the bam file? I know deeptools, bamcoverage can do, but the output is bigwig file which can not be used in seqminer.

Thanks!!!

Jianhuang

[Tao]

Hi Jianhuang,

If you think a normalisation with the total read number is enough, you can do a random sampling of your samples to get the same number of reads for each sample. You can use for example macs2 randsample command.

nowadays I will do the sampling for most of the projects to 20 million reads. This can also accelate the loading of the data and reduce the memory requirement.

best,

Tao

在 2018年4月18日星期三 UTC+2下午4:29:53，jianhuang lin写道：

[jianhuang lin]

Thanks a lot!! This really helps!

在 2018年4月18日星期三 UTC-4上午10:29:53，jianhuang lin写道：