counting the reads that mapped to a part of transcriptome

Question

[Sara]

I have RNAseq data and trying to count the reads that mapped only to a part of 5'UTR (in fact the whole 5'UTR except the first 50 nts). do you know how to do that?

Answer 1

[tarek.mohamed]

Hi Sara,

Did you align the raw reads to a reference genome? if so what is the current format of your aligned reads?

Tarek

[Sara]

I can actually align to the part of transcriptome that I want (first I have to make fasta file and use it as reference file to align to). I can also make gtf file using the same file and count using that. but in both cases it returns 0 counts. so I thought maybe I should align to the genome and count on the reads that mapped to that part. but here I do not know how to do that.

Answer 2

[glihm]

Hello Sara,

once you have mapped the data against the genome of your interest, you can simply use FeatureCounts program for instance:

featureCounts -a yourGTF.gtf -t UTR5 -g transcript_id -o counts_UTR5_transcripts.txt

The -t option allows you to select a type of features to be counted. The -g option allows you to choose "how to group your data". Using transcript_id (or something equivalent in your GTF/GFF), you tell to FeatureCounts: "ouput the counts for each transcript".

[tarek.mohamed]

Also if you prefer to use R, you can use "Rsubread" package to count your reads using featureCounts () as mention by glihm.

T