exon quantification for tissue samples

Question

[Varun Gupta]

Hi Everyone,

I have some tissue RNA-SEQ samples and I have mapped it to hg19 ref genome to get the bam files. I am trying to get the exon counts for a particular exon for all those tissue samples from the bam files. This exon is not annotated in the gif/gff file. I have the coordinates of this exon and have made an exon.bed file which just has one entry which is my exon coordinate. I am trying to use various methods to get exon count for this exon, but they don't seem to give me accurate results.

I will explain a bit about the exon. My exon is considered to be a micro exon and therefore the read splits into more than one place. Sometimes this exon is completely skipped and hence we have alternative isoform.

When I try to use multiBamCov, I am getting those reads also counted which is skipping this micro exon completely. This is because the upstream 5' ss is the same for this micro exon and other exon down stream. I want to count those reads which are

contained within the exon
some part of the read overlaps with the exon as well as 
all those splice junctions count which include this exon

How can I do this. Hope to hear soon.

Thanks

Varun

Answer 1

[EagleEye]

Why don't you try with standard quantification tools like featureCounts or HTseq. All you have to do is insert your extra exon for particular gene/isoform in standard GTF file containing all the genes and you will get the counts for your gene of interest including your new exon or exon level (using featureCounts).

If you want just the count for your exon of interest, prepare your SAF file (just simple as bed file) and use featurecounts.

[Varun Gupta]

Hi, I thought of inserting the exon in the GTF file. I will insert it and see what I get for my exon. I should use featureCounts since I have a bed file for the exon already. I will give it a try and let you know

Thanks