alternative splicing analysis question

[Dariel Ashton-Beaucage]

Hi,

I'm considering using your program to analyze my RNA-seq data. In addition to identifying alternate junctions and comparing the changes in AS across samples, I would also like to be able to be able to distinguish unspliced (pre-mRNA) from spliced forms. My samples have a significant amount of pre-mRNA as RTs were done using random primers. Does the AltAnalyze software take this into account, or would it confuse unspliced pre-mRNA with intron retention?

Thanks,

Dariel
_____________________________
Dariel Ashton-Beaucage
IRIC - Université de Montréal
(514) 343-6111 ext.8092
d.ashton...@umontreal.ca

[Nathan Salomonis]

Hi Dariel,

Great question. For RNA-Seq data exons and junctions are both analyzed
in parallel pipelines with the results merged at the end. What this
gives you are competing junctions that detect alternative expression
of an exon(s) along with independent confirmation for regulation of
that exon independent of the junction expression (relative to the
gene/transcription model for that gene). For unspliced pre-mRNA, this
should be detected as intron retention, in fact, introns are also
evaluated for alternative regulation using the splicing-index method,
the same as exons (e.g., an intron is upregulated relative to the
constitutive exons). This should be called as intron retention,
assuming you have more or less unspliced RNAs in a given state.

What you won't get is a model that substracts out the expression of
unspliced transcripts versus spliced transcripts, except for the
junction based analyses which will only consider the expression of
spliced transcripts. But I wouldn't expect the direct detection of a
known or novel exon to be impacted by pre-mRNA in a confusing way.

You may be able to see the overall contribution of intron retention
with AltAnalyze's new QC methods, but no deconvolution methods to
remove these effects.

Best,
NS

> --
> You received this message because you are subscribed to the Google Groups
> "Alternative Splicing and Functional Prediction" group.
> To post to this group, send email to alt_pre...@googlegroups.com.
> To unsubscribe from this group, send email to
> alt_predictio...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/alt_predictions?hl=en.