New at RNA-seq

[jpelletier]

Hello Everyone,

I am new at RNA-seq, switching from the good old Genechip to this technology... We have very limited RNA input amounts to play with and since we've been happy with Nugen amplification system in the past, so we're planning on using the 3'DGE amplification kit.

My questions are in regard of Nugen's Encore NGS library prep kit. We're going to do single-end sequencing for differential gene expression analysis. We have a ref genome so we don't really care about gene coverage and such. In addition, I received a bit of conflicting advice about cDNA fragmentation, or the need to do it. Any thoughts on the matter?

thanks!

[TUCF JSS]

I think the beauty of the Encoure NGS kit for RNA Seq is that it can take fragmented RNA and create cDNA from it. So if this is the case then no cDNA fragmentation is necessary (perhaps size selection, but not fragmentation). However if your fragment sizes of cDNA are not ideal for your sequencing system (too big) then fragmentation is necessary.

I hope this helps with your question, but I wasn't terribly sure based on your post if this is what you meant!

~Jimmy

[jpelletier]

Thanks for your reply Jimmy! I realized after posting that my question was a bit unclear... sorry.

I think I will definitely have to reduce the size of the cDNA population to optimize cluster formation since the cDNA size range I get is ~100-1000nt, centered on 200nt. What do you think? (We'll be using Illumina.)

thanks!
Julie

[TUCF JSS]

Well if you're already seeing a general clustering around the 200bp region, then you may not want to fragment the cDNA further, as that can lead to potential loss of DNA. In our lab, before switching to the Covaris, we used nebulization. This would often leave us with a very wide range (from 100-1000, like yours) centered around 400. But as long as we began the library prep process with 1ug or more of the DNA we were fine. We selected our bands by running our ligated products on a gel and taking a very small fragment.

The only other thing to consider is what type of RNA you had initially dealt with. Do you know if it was already sheared/randomized? You don't want to accidentally favor 1 type of transcript that happens to be at a certain size or simply sequence a bunch of ribosomal RNA either. If you show a bioanalysis report or even a gel pic (maybe even just a rough description) to a NuGen tech over the phone they can tell you exactly how to handle your genome! They are extremely helpful. And no, I do not work for NuGen

[jpelletier]

The tech support at Illumina was quite adamant I shouldn't have anything above 600nt, it would otherwise hinder the identification of clusters. (Ditto for the in-line barcoding Nugen uses instead of using a second primer to read the index like Illumina's Truseq system...)

Our RNA is quite a bit compromised, it comes from Laser-microdissected tissues. We can see a bit of rRNA peaks but it's mostly a big smear. We assume it's random shearing...

I'll talk to Nugen again, you are right they are usually really helpful!
thanks again!

[TUCF JSS]

You definitely don't want anything in your final sequencing product to be above 600! That is true. However if you are just starting the library preparation process you can use all the DNA you have and do a size selection at a later point (IE Gel selection, Ampure Beads, etc). Unless of course you have a covaris or a very reliable method of shearing. Otherwise I would use what you have right now!