Differential analysis of non coding RNA seq

[Claudia34]

Hello ,

I'm working on ncRNA seq libraries obtained with Illumina... I just start to have a look about statistical analysis between two libraries : for example, I would like to know if the microRNA 1 is significantly different in the library 1 comparing to the library 2...
I am not really aware of the kind of analysis I could perform to analyze this type of difference...

Does anyone has any idea about this ? Any publication to advise ???

Thank you,

Claudia

[Simon Anders]

Hi Claudia,

you can use the same tools for miRNA as is used for normal mRNA. Currently, there are three tools available to test for differential expression:

- edgeR (Robinson, McCarthy, Smyth)
- DESeq (Anders (=me), Huber)
- baySeq (Hardcastle et al.)

Although it is often mentioned here, I do not list cuffdiff (part of the cufflinks package by Trapnell et al.) as it does not test for association of a detected difference with an experimental condition.

In your case, you cannot do any proper analysis, of course, as you don't have replicates. (See e.g. the relevant posts in this thread.)

Simon

[chebel]

Hi Claudia,

There are quite a few web-based tools available for analysis of deep sequencing small RNA data.

Here is a good list of them.

Chris

[lpachter]

Dear Claudia,

Cuffdiff can be used to test for differential expression, and you may wish to try it. In particular, the new version can be used with replicates and also supports bias correction (performed individually on each replicate). Version 0.9 is downloadable from

http://cufflinks.cbcb.umd.edu/

[honey]

micro-RNA

Hi Ipachter,

If I have to analyze microRNA What specific step I need to change. Just use the annotation file for micro_RNA and use total RNA for RNA-seq? that should be it or I need to do some more changes. I am very interested in using cufflinks.

Thanks.

[ae_ucla]

Has anyone used Cufflinks for microRNA data?