strand specific RNAseq

[Pepe]

Hi all,

I am interested in hearing from users preparing strand-specific RNAseq libraries. We work with plants.

We are considering the ScriptSeq kit, but we are not sure about how well does it work with plant tissues.
Anyone has a protocol that can be used in combination with the Illumina Truseq kits?
Any other alternatives?

I'd really appreciate any information. Thanks in advance.

[kmcarr]

No experience with the ScriptSeq kit but I will relate our experience with strand specific RNA-Seq.

We have recently been working with some ssRNA-Seq data from a non-model plant. The library was prepared using the Illumina RNA-ligation method. We found that the depth of read coverage across contig assemblies was extremely uneven, by >2 orders of magnitude in some cases. It seems that this is a typical outcome with this library prep method.

Check out this paper by Levin et al. "Comprehensive comparative analysis of strand-specific RNA sequencing methods." Nat Methods (2010) vol. 7 (9) pp. 709-15 (Link). They examine several methods of preparing ssRNA-Seq libraries. They did not test the ScriptSeq kit themselves but it appears that the ScriptSeq method would falls into the "not so random" or "not not so random" class. This method did not fare well in their analysis. The best method by their estimation is the dUTP second strand method, with the Illumina ligation protocol a close second (despite the unevenness problems noted above). In our core we are looking at implementing the dUTP protocol for future ssRNA-Seq preps.

[Pepe]

Thanks for your reply.
We have decided to do the same, the hope is that we can integrate the new Truseq kits (high throughput) and the dUTP methods (best quality according to Levin et al.).

Please feel free to PM us or write here to give us any advice you have on how to do this.
I'll be glad to keep you posted about our developments, although it will take some time to set up everything here.

[upendra_35]

I have used ScriptSeq and though i can detect reads from antisense strand but the protocol suffers from very biases such as 3' end and G bias at the beginning of the read. Also note that it is not a high throughput protocol and is very expensive per library. Hope this helps

[upendra_35]

Thanks for your reply.
We have decided to do the same, the hope is that we can integrate the new Truseq kits (high throughput) and the dUTP methods (best quality according to Levin et al.).

Please feel free to PM us or write here to give us any advice you have on how to do this.
I'll be glad to keep you posted about our developments, although it will take some time to set up everything here.

hi pepe

Just wondering have you figured out how to integrate the TruSeq kit for the strand specific libraries?

Thanks in advance.....

[perrinwang]

http://dx.plos.org/10.1371/journal.pone.0026426