miRNA mature and star sequences, isomiRs etc

[naluru]

Hello,

I have asked this question before and didn't get any replies. I thought I will post it again as many others might be having similar problems.

I am working with miRNA-seq data set and I mapped them to the genome after adapter trimming. I am able to count the known miRNA reads using genome coordinates of precursor miRNA. That worked perfectly! Thanks to many of you who helped me do that.

What I am trying to do now is to get the read counts for individual mature miRNAs and star sequences. I am having hard time getting that information in a simplistic way.

Is there any way I can process the sam or bam file to get all the individual mature and star sequence counts separately? Right now, I am using IGV to view the alignment and basically counting the number of reads aligned to mature and star sequence. There got to be a much clever way to get this information.

Also, is there any way to detect isomiRs?

Any comments/suggestions are highly appreciated.

Thank you

[zee]

You could try something like coverageBed that is part of the bedTools package e.g.

coverageBed -abam samplex.bam -b mature.[bed|gff] > mature.samplex.coverage

Repeat the same process for your star miRNAs. You could also just work with some naming convention to distinguish your stars from mature miRNAs.

The last three columns of mature.samplex.coverage gives your depth counts and other useful information.

Hope this helps.

[volks]

I am able to count the known miRNA reads using genome coordinates of precursor miRNA.
...
What I am trying to do now is to get the read counts for individual mature miRNAs and star sequences.

use the genome coordinates of individual mature and star sequences. or did i not understand the question?

[naluru]

Thank you for both the suggestions.

Thank you, zee. I will try bedTools.

Volks, you are correct and I thought about it. The GFF files in miRBase website have the coordinates only for precursor sequences and not to mature/star sequences separately. Also, not all the star sequences are annotated in miRBase. I am seeing many reads that are mapped to star sequences (unannotated).

Please let me know if you think of any way to deal with this? Thanks, Neel