Dear Community,
I am new to ChIP-seq so please be gentle.
Analyzing ChIP-seq data - public and from my collaborators - I realized that the minimum of reads is varying very much - NCBI's minimum to submit ChIP-seq is 1 million short reads I heard and I saw published data with even less than that.
Looking at the distribution of reads in UCSC browser (by uploading the .bam files) it looks very random over the whole genome, sometimes reads are stacking up a bit.
Here the questions:
A) Should you normally be able to spot peaks visually?
The peak callers take into account fragment size as well I suppose, but usually should you be able to see distinct peaks by eye?
B) If not - if it looks all over the genome, does that mean that there are not enough reads to detect peaks reliable?
C) How many reads would you recommend minimum for Illumina GAIIx - read length 37 - 42nt for TFs like Myc-n or NFkB-p50?
Thank you very much
I am new to ChIP-seq so please be gentle.
Analyzing ChIP-seq data - public and from my collaborators - I realized that the minimum of reads is varying very much - NCBI's minimum to submit ChIP-seq is 1 million short reads I heard and I saw published data with even less than that.
Looking at the distribution of reads in UCSC browser (by uploading the .bam files) it looks very random over the whole genome, sometimes reads are stacking up a bit.
Here the questions:
A) Should you normally be able to spot peaks visually?
The peak callers take into account fragment size as well I suppose, but usually should you be able to see distinct peaks by eye?
B) If not - if it looks all over the genome, does that mean that there are not enough reads to detect peaks reliable?
C) How many reads would you recommend minimum for Illumina GAIIx - read length 37 - 42nt for TFs like Myc-n or NFkB-p50?
Thank you very much
Comment