Analysis of Gro-seq data

[sikidiri]

I want to analyse a published gro-seq data. But I am new to it.
I want to see the direction of transcripts. For this, do I need to separate the positive and negative tags and then map them on to the human genome?
Please advise me.
Thanks.

[steven]

I'd say that the "positive" or "negative" strand information is precisely supposed to come from the alignment.

[sikidiri]

Thanks for you reply. I have already the aligned reads in bed format. I want to see these tags on the human genome e.g at the TSS and want to see the direction of the tags. Do I need to segregate the sense and antisense genes or shall I segregate the tags based upon the strand?

[steven]

if there is a strand information specified in your BED ("+" or "-" in the 6th column) then it should be easy to compare the reads with some gene annotation. The idea would be to compare
- the positions of the reads vs. the positions of the annotated exons to assign reads to genes
- the strand info of the reads vs. the strand of the corresponding gene to assign a "sense" or "antisense" label to each read
Look e.g. for "BEDtools" (and send me a PM if you don't find)

[sikidiri]

Hello,
0


I have two chip-seq samples for the same protein in embryonic stem (ES) cells and rationic acid induced cells. I have obtained around 800 peaks in ES cells and around 7500 peaks in induced cells. Protocol, antibody, peak calling paramteres (MACS) and the person who has done the the experiments are all same. Number of reads obtained in both the samples is similar with similar level of background. If I see peaks in my new dataset, it has good enrichment as compared to the old one at the same region (~50% higher enrichment). I want to know, is this the real biological difference or because of deep sequencing, in the new data set I see good enrichment of tags which is not seen in the old dataset. How to rule out any technical problems, if there are any? Any suggestions are most welcome. Thanks.