How do I identify splicing differences from RNA-Seq

[adrian]

Dear group,

I think I am back to square 1 on this issue.

I have RNA-Seq data from 3 samples.

I used tophat / cufflinks pipeline.

From PCR experiment I know that a Gene xyz (that has only 1 known transcript) is spliced at exons 4, 5 and 6. In sample 1, exons 4-6 are joined and also have transcripts 4-5-6.

In sample 2, I know exon joining 4-6 is not happening. Only one form 4-5-6 happening.

In sample 3, there should be both 4-6 and 4-5-6 exon join.

My question:

1. From various cufflink outputs, how do I find this splicing differences in samples 1, 2 and 3.

2. How do Identify other genes that behave the same way.

3. How can I visually observe the identified genes where splicing is happening.

Could any one please help me this. I am desperate and need help.

I just cant wrap my around this. If cufflinks is not designed to answer these questions, what other softwares could help me.

thanks

[adrian]

sorry again to bump another question by me.

I see over 100 hits, and no answers.

I am wondering if my question is really stupid or a basic question that is obvious or my question don't make sense. I appreciate also any reply stating that or pointing me to a resource where I can find the answer.

Thanks
Adrian.

[kopi-o]

No, it's not a stupid question, these issues are not straightforward.

From Cufflinks output, you could use something like ASProfile and go from there. From Cuffdiff, you could take a look at CummeRbund and see if that fits.

Check out this list of related software:

http://www.rna-seqblog.com/data-analysis/expression-tools/methods-to-study-eventisoform-expression-and-alternative-splicing-from-rna-seq/

[adrian]

Thank you very much for your reply.

it is frustrating to sift through 400 genes from isoforms.diff file and identify what exon arrangement is different between isoforms of each gene.

Currently I am lining up junctions.bed and bam file in IGV which is a pain in eyes and confusing.