Alternative splicing analysis following kallisto?

Question

[dina.hesham139]

Hi,

I used Kallisto for analyzing Rice RNAseq data, is there any tool that I can use to study alternative splicing following kallisto, a tool that doesn't require bam files and supports Rice genome?

Thanks in advance!

[lessismore]

Hey! So far you are the only one using this approach. I didnt find a single paper mapping RNAseq data against rice transcriptome. Thats pretty curious

Answer 1

[Istvan Albert]

Since with Kallisto you are quantifying against transcripts hence you can go ahead with your analysis.

If you trying to discover new splice variants you would need to align against the genome with a different method.

[dina.hesham139]

Thanks alot Istvan ! What other pipeline would you recommend for studing alternative splicing? esp. for rice and in my case I have a condition that doesn't have a replicate (i.e. I can't use rMATS for example!)?

[joomi]

Hey Dina.hesham139 can I ask you what you ended up using? I'm using Kallisto also and Sleuth, but then someone mentioned that Sleuth might not be the best method. I was wondering what you used?

Answer 2

[Kristoffer Vitting-Seerup]

Hi Dina

Take a look at my R package IsoformSwitchAnalyzeR (you need version 1.1.07 or higher) - it directly supports Kallisto data and can analyze alternative splicing both individual genes as well as genome wide level. See examples for individual genes (isoform switch) here or for the global analysis see this part of the vignette.

And as Istvan suggest if you are interested in novel splicing you need a different analysis pipeline.

[dina.hesham139]

Hi kristoffer, I 'm trying to analyze my kallisto data with IsoformSwitchAnalyzeR but I get this error when I use the importRdata function, I get this error:

> aSwitchList <- importRdata(
+   isoformCountMatrix   = mySwitchList$counts,
+   isoformRepExpression = mySwitchList$abundance,
+   designMatrix         = designMatrix,
+   isoformExonAnnoation = "/rice/Oryza_sativa.IRGSP-1.0.37.gtf",
+   showProgress = T)
Step 1 of 6: Checking data...
Step 2 of 6: Obtaining annotation...
    importing GTF (this may take a while)
converting annotated CDSs
Error in importRdata(isoformCountMatrix = mySwitchList$counts, isoformRepExpression = mySwitchList$abundance,  : 

The annotation and quantification (count/abundance matrix and isoform annotation) seems to be different (jacard similarity < 0.95). 
Either isforoms found in the annotation are not quantifed or vise versa. 
Specifically:
 42346 isoforms were quantified.
 98663 isoforms are annotated.
 Only 42346 overlap.

This combination cannot be analyzed since it will cause discrepencies between quantification and annotation thereby skewing all analysis.

Please make sure they belong together and try again. For more info see the FAQ in the vignette.

knowing that I use the ensembl gtf annotation file that used in kallisto analysis.

Any Help?!

thanks in advance

[Kristoffer Vitting-Seerup]

Hi Dina

Sorry for not seeing this earlier. Did you solve this problem?