Illumina directional RNA-seq protocol

[Herve]

Hi,
I am thinking about using Illumina's strand specific RNA seq protocol, based on the use of their small-RNA library prep kit (see attachment).
Having strand specificity is a great deal, but I am concerned as I have not seen anybody using it yet and I am wondering whether there is a good reason for that. So:
Has anyone tried it?
Or any concern about the way it proceeds?
Thanks!

Herve

[der_eiskern]

it appears the Sanger Institute has been performing ssRNA-seq (published July 2009)

[Herve]

Thanks a lot for the reference.
I was not aware of it.
By looking at the paper it however seems that the authors have some doubts about the use of such a method for quantification purposes, as the correlation with microarrays results seems rather low.
I however have a hard time figuring out why the protocol might be responsible less quantitative than the non strand-specific classic Illumina library generation kit...

[das]

I tried the protocol and it worked as advertised.
It compared well with traditional RNAseq protocols which I did in parallel.

[scooter]

I would not be too surprised that the Seq results to not match the array results very well. In my experience this is a universal phenom, regardless of the library prep method used. I think it is due to a number of variables, but one is likely to be simply that the dynamic range is so different between the two platforms. In addition, comparing a hybridization approach to collecting individual tags are very, very different modes of quantitation, so it is like comparing apples to grapefruits.

Now, as to which one is more 'correct'? That is a very different question.....

I have not used the illumina method. However, I do have a lot of experience with the SOLiD directional RNAseq method and can say that it works quite well and produces very quantitative data, so I imagine the illumina method is not so different in result even if the method is a bit different.

[fennan]

I have a dummy question...

Where is strandness information of each read specified in the data produced by the sequencer?

Thanks in advance!
Fernando

[gaster]

It is worth pointing out that the Sanger paper listed above does not use the same ss-mRNAseq method used in the Illumina protocol. Also, in the method the Sanger center uses in that paper, there are details of the library construction that are not clearly understood and remain oddly speculative. The method is described in more detail here:

http://nar.oxfordjournals.org/cgi/co...bstract/gkp811

The Sanger Center has a new directional method described here:

http://www.nature.com/nmeth/journal/v7/n2/full/nmeth.1417.html

Good Luck!

[das]

Strand-information is preserved by ligation of specific adaptors to RNA. This is why protocol is similar to small RNA sequencing procedures.

[link1]

any advice on which one works better? mRNAseq or directional protocol?

[das]

Directional vs traditional

Well, both protocols worked for me. Obviously directional protocol provides user with more information than traditional, but directional protocol gives more uneven (choppy) coverage along the genes. So depending on your goals, that may prove to be a disadvantage.

Das

[link1]

thanks Das, thats good to hear. I'm going to try both protocols next week. Will let everyone know how it goes.