RNA-seq: directional vs. non-directional

[frefra]

Hello all!

This is my first post here! But I am not just new at the forum, I am also new on the field of RNA-seq. So please be lenient with me! (-; I already read a lot in this nice forum and many things gets clearer to me now. But at the moment I still have a question. I am sure somebody can answer…

We want to perform a whole transcriptome sequencing experiment using the Illumina HiScanSQ system. Samples will be obtained from fish and tapeworms. For the fish species an annotated reference genome is available. At the moment collogues are working on a reference genome for the tapeworm but I am not sure if it will be annotated till we need it.

We are now at a point where we have to decide for the library prep procedure. We want to isolate total RNA and analyze only mRNA – that is clear. What we are not sure about is whether we should perform directional (strand-specific) or non-directional sequencing.

I already tried to find out the advantages and disadvantages of both methods but I did not get satisfying answers. Could please somebody explain me the benefits of both methods and how the methods differ concerning data evaluation.

Thanks a lot!!!

Cheers

frefra

[dpryan]

In cases where it's unclear how good the annotation is you're probably best off using a strand-specific protocol. I would argue that this has the advantage of making feature (genes in this case) discovery a bit easier, since you won't have overlapping genes merged together). This is a somewhat naive recommendation, of course, since I don't know much about the 2 genomes in question.

Edit: To also answer the remainder of your post...

The non-stranded protocol at least used to be cheaper (and presumably easier, though I just have a core facility make the libraries for me!). For those of us working on well characterized species (e.g., human or mouse), the annotations are already good enough that we can get away with the cheaper/quicker methods. If I were working on your project, however, I'd be more confident moving forward with the directional library prep since there are a lot of unknowns in the annotation/reference (you may need to assemble the transcriptome at least partly, in which case many relevant assemblers can use the strand information).

[frefra]

Hi dpryan,

Tanks a lot for your fast and clear answer! If a directional library prep makes life easy afterwards I am sure I am able to convince my supervisor to spend a little bit more money for the library prep! (-;

Cheers!

frefra

[karthick.n]

Can anyone of you tell me how to find the method of RNAseq (strand specific/Un-stranded) using IGV tool. i have the sample sequencing data. But i don't know the method of library preparation they did.

Thanks in Advance!

Cheers,
Karthick

[cmbetts]

Color by first strand in pair and pick a housekeeping gene (actin, GAPDH, etc). If you see roughly equal color mix, nondirectional. Nearly all the same color, directional with the protocol determined by whether the pairs are sense or antisense to the ORF.

Some of the counting tools also output both directional and nondirectional counts at the same time, which would tell you without the need for visual inspection while getting you a step closer to analysis

[karthick.n]

Thank you so much for the response!

[oceanapart technologies]

smart home, wifi and internet, mobile and tablets, tv mounting, audio video.
Oceanapart Technologies