SEQanswers

Go Back   SEQanswers > Applications Forums > RNA Sequencing



Similar Threads
Thread Thread Starter Forum Replies Last Post
Problem with high background in some RNA-seq samples pmcget RNA Sequencing 10 11-27-2012 08:09 AM
RNA-Seq from FFPE Samples gavin.oliver General 2 04-30-2012 12:14 AM
RNA-seq Galaxy workflow for PE barcoded samples? jjw14 Bioinformatics 0 04-19-2011 12:58 PM
New low-input rRNA removal kit, ideal for FFPE samples epibio Vendor Forum 0 11-04-2010 09:55 AM
New! Ovation WGA FFPE and RNA-Seq FFPE workflow solutions from NuGEN Technologies NuGEN Vendor Forum 0 11-03-2010 09:02 AM

Reply
 
Thread Tools
Old 04-27-2012, 04:49 AM   #1
SallyH
Junior Member
 
Location: UK

Join Date: Apr 2012
Posts: 4
Default RNA Seq from FFPE samples

Has anyone done any RNA Seq analysis from FFPE samples and if so can anyone offer any advice please on either the use of the mRNA-Seq Sample Preparation Kit or DSN normalization?

Any advice would be greatly appreciated!
SallyH is offline   Reply With Quote
Old 04-27-2012, 05:44 AM   #2
kmcarr
Senior Member
 
Location: USA, Midwest

Join Date: May 2008
Posts: 1,177
Default

Quote:
Originally Posted by SallyH View Post
Has anyone done any RNA Seq analysis from FFPE samples and if so can anyone offer any advice please on either the use of the mRNA-Seq Sample Preparation Kit or DSN normalization?

Any advice would be greatly appreciated!
We have not done any FFPE samples but I'll offer advice none the less. The standard mRNA-Seq library prep uses a polyA selection to remove rRNA; it assumes that your total RNA sample is very high quality (i.e. not degraded) so that you are pulling out full length mRNAs. FFPE samples are highly degraded so the standard mRNA library prep is not appropriate.

When Illumina originally reported the DSN protocol they highlighted using it with FFPE samples since it works even degraded RNA. That is the way to go for your sample.
kmcarr is offline   Reply With Quote
Old 04-27-2012, 05:54 AM   #3
ETHANol
Senior Member
 
Location: Western Australia

Join Date: Feb 2010
Posts: 308
Default

I'm making some FFPE RNA-seq libraries right now.

The RNA was purified with the Qiagen FFPE RNeasy minElute kit.

For the libraries, I'm using the TruSeq RNA kit, except I've replaced the poly A selection with Ribo-Zero depletion. I did a test run of the protocol with one sample yesterday and it more or less looked ok on the Bioanalyzer. We'll see in some weeks what the sequence data looks like.
__________________
--------------
Ethan
ETHANol is offline   Reply With Quote
Old 04-27-2012, 08:16 AM   #4
kmcarr
Senior Member
 
Location: USA, Midwest

Join Date: May 2008
Posts: 1,177
Default

Quote:
Originally Posted by ETHANol View Post
I'm making some FFPE RNA-seq libraries right now.

The RNA was purified with the Qiagen FFPE RNeasy minElute kit.

For the libraries, I'm using the TruSeq RNA kit, except I've replaced the poly A selection with Ribo-Zero depletion. I did a test run of the protocol with one sample yesterday and it more or less looked ok on the Bioanalyzer. We'll see in some weeks what the sequence data looks like.
I don't do the wet lab stuff myself, meaning my knowledge here is theoretical so take it for what it's worth, but I question the effectiveness of Ribo-Zero on FFPE materials. Ribo-Zero, or similar rRNA depletion kits depend on targeted binding of specific oligos to pull down rRNA. If the rRNA is fragmented then only those fragments complementary to the oligos will be depleted, all of the other fragments will still be present in your library.

This was one of the main advantages of the DSN method for rRNA depletion, it does not depend on any specific sequences of the RNA you are trying to deplete. It is solely based on concentration, and the relative concentrations of rRNA and mRNA should remain the same whether or not the sample is degraded.
kmcarr is offline   Reply With Quote
Old 04-27-2012, 08:38 AM   #5
ETHANol
Senior Member
 
Location: Western Australia

Join Date: Feb 2010
Posts: 308
Default

That is what I am most worried about, but this what epi-bio advertises:

Quote:
The Ribo-Zero™ rRNA Removal Kits (Human/Mouse/Rat) remove more ribosomal RNA (rRNA) from intact RNA, partially degraded RNA, and RNA isolated from formalin-fixed paraffin-embedded (FFPE) human, mouse, and rat RNA preparations than any other method. The Ribo-Zero kits set a new standard for rRNA removal for whole-transcriptome RNA-Seq.

Benefits

Removes >99.9% of 28S, 18S and 5.8S rRNA, and >95% of 5S rRNA from intact, and partially degraded, including FFPE-derived, human, mouse, and rat RNA preparations (Tables 1 and 2).
Single-pass, 90- to 120-minute process yields RNA sufficiently depleted of rRNA for RNA-Seq (Table 2), random-primed cDNA synthesis, and other applications.
Recovers mRNAs and ncRNAs for whole-transcriptome RNA-Seq.
Preserves the transcript profile of the sample (Fig. 1 and 2).
Recovers both the 5´- and 3´-end fragments of mRNAs and ncRNAs in a partially degraded RNA sample, including FFPE RNA.
And experimentally backed hype:
http://epicentral.blogspot.com/2010/...q-library.html

One thing is clear from what I have read is Ribominus form Invitrogen/Life Tech does not work as advertised. Ribo-Zero has gotten good reviews.
__________________
--------------
Ethan
ETHANol is offline   Reply With Quote
Old 04-27-2012, 08:41 AM   #6
kmcarr
Senior Member
 
Location: USA, Midwest

Join Date: May 2008
Posts: 1,177
Default

Thanks ETHANol, I happily stand corrected.
kmcarr is offline   Reply With Quote
Old 04-27-2012, 09:14 AM   #7
mnkyboy
Member
 
Location: Seattle, WA

Join Date: Mar 2009
Posts: 87
Default

I have done it both ways using Total RNA-seq bypassing the poly-A step with the TruSeq kits.

In my hands the RiboZero works extremely well, but DSN can work just as well but if you are not experienced with it there is a higher chance of problems. We chose to use RiboZero for our FFPE studies due to its ease of use and reproducible reduction in rRNA.

Also there is some debate whether to fragment or not. I chose to fragment to achieve uniform fragment size and we got very good 5'-3' coverage. We ended up doing 2x50bp and treated these as we would for degraded RNA samples for total RNA-seq.
mnkyboy is offline   Reply With Quote
Old 04-27-2012, 09:18 AM   #8
mnkyboy
Member
 
Location: Seattle, WA

Join Date: Mar 2009
Posts: 87
Default

Also to echo ETHANol I have used both RiboMinus and RiboZero and there is not comparison. In fact RiboMinus should not even be an option they don't even touch the performance of the RiboZero beads for an rRNA reduction.
mnkyboy is offline   Reply With Quote
Old 04-27-2012, 01:28 PM   #9
NextGenSeq
Senior Member
 
Location: USA

Join Date: Apr 2009
Posts: 482
Default

You'll get a lot of rRNA in your libraries using degraded FFPE RNA.
NextGenSeq is offline   Reply With Quote
Old 04-30-2012, 12:53 AM   #10
JakobHedegaard
Member
 
Location: Aarhus, Denmark

Join Date: Mar 2008
Posts: 57
Default

Hi,
We have done RNA-Seq on RNA from FFPE using Ribo-Zero GOLD in combination with ScriptSeq (both from Epicentre).
On 240 months old FFPE samples we find 3-6 % of the reads mapping to rRNA - and much less (<1%) in younger FFPE samples.
Cheers,
Jakob
JakobHedegaard is offline   Reply With Quote
Old 07-10-2012, 07:07 PM   #11
lahoman
Member
 
Location: Houston

Join Date: Jan 2011
Posts: 12
Smile FFPE RNA using TRueseq RNA kit

Thanks for sharing your experience with FFPE RNA using TRueseq RNA kit.

For some specific reasons, we have to choose DSN-normalization method for my FFPE RNA library prepared with Trueseq RNA kit. Illumina told me that the and the DSN protocol won't work well with library prepared with Trueseq RNA kit.

Could you share more details, like input amount of adapter-ligated and enriched DNA, primers and condition used for final amplification?

Thanks in advance.
lahoman is offline   Reply With Quote
Old 07-11-2012, 03:45 AM   #12
HedleyC
Junior Member
 
Location: Nr. Manchester, UK

Join Date: Nov 2008
Posts: 5
Unhappy Re: Ribozero Gold and ScriptSeq on FFPE

Quote:
Originally Posted by JakobHedegaard View Post
Hi,
We have done RNA-Seq on RNA from FFPE using Ribo-Zero GOLD in combination with ScriptSeq (both from Epicentre).
On 240 months old FFPE samples we find 3-6 % of the reads mapping to rRNA - and much less (<1%) in younger FFPE samples.
Cheers,
Jakob
Jakob, I have been involved in a piece of work trying to do the same thing - but we just got massive adapter contamination in the final data. What size range was your FFPE RNA peak after RiboZero GOLD but before the ScriptSeq? Ours was between 25-500bp approximately, peak generally around 100.

Regards,
Hedley
HedleyC is offline   Reply With Quote
Old 07-11-2012, 11:31 AM   #13
lahoman
Member
 
Location: Houston

Join Date: Jan 2011
Posts: 12
Default FFPE samples with Trueseq RNA kit

I am working on the RANseq with FFPE RNA samples. I just noticed that the size of my dscDNA sample ranged from 100-200bp while after the end-repaire, A-tailing and enrichment, the size of my liberary ranged from 200-1000bp. Does anyone can tell me the reason and why is good for loading into the sequencer?

Thanks in advance!
lahoman is offline   Reply With Quote
Old 07-11-2012, 12:00 PM   #14
ETHANol
Senior Member
 
Location: Western Australia

Join Date: Feb 2010
Posts: 308
Default

Iahoman -- probably because your libraries are over amplified. There are a few threads on this topic here. Do a search. I do this to get the number of cycles right:
http://ethanomics.wordpress.com/ngs-...tion-protocol/

A completely different question...
What are people doing for fragmentation of RNA from FFPE samples? Do you decrease the incubation time at 94˚C? Do you fragment at all? Any thoughts on this.
__________________
--------------
Ethan
ETHANol is offline   Reply With Quote
Old 07-12-2012, 07:24 AM   #15
lahoman
Member
 
Location: Houston

Join Date: Jan 2011
Posts: 12
Default FFPE RNAseq

Thanks, Ethanol!
lahoman is offline   Reply With Quote
Old 07-12-2012, 12:48 PM   #16
lahoman
Member
 
Location: Houston

Join Date: Jan 2011
Posts: 12
Default

I did not perform Fragmentation for my 1st FFPE sample. Now I am thinking I may need do to do that to narrow the insert size ...
lahoman is offline   Reply With Quote
Old 07-31-2012, 04:02 AM   #17
JakobHedegaard
Member
 
Location: Aarhus, Denmark

Join Date: Mar 2008
Posts: 57
Default

Quote:
Originally Posted by HedleyC View Post
Jakob, I have been involved in a piece of work trying to do the same thing - but we just got massive adapter contamination in the final data. What size range was your FFPE RNA peak after RiboZero GOLD but before the ScriptSeq? Ours was between 25-500bp approximately, peak generally around 100.

Regards,
Hedley
RNA size range as yours - but with a sligthly higher peak ~200 nt. How do you purify the library after amplification? Using MiniElute will give you some "primer-dimer" contamination. We are using AMPure XP but you may need to play around with the bead ratio to get rid of the "primer-dimer" without removing too much of the library. Have you considered gel purification?
/Jakob
JakobHedegaard is offline   Reply With Quote
Old 07-31-2012, 06:29 AM   #18
HedleyC
Junior Member
 
Location: Nr. Manchester, UK

Join Date: Nov 2008
Posts: 5
Default

Quote:
Originally Posted by JakobHedegaard View Post
RNA size range as yours - but with a sligthly higher peak ~200 nt. How do you purify the library after amplification? Using MiniElute will give you some "primer-dimer" contamination. We are using AMPure XP but you may need to play around with the bead ratio to get rid of the "primer-dimer" without removing too much of the library. Have you considered gel purification?
/Jakob
Hi Jakob,
The ScriptSeq and later steps were performed by a 3rd party. They used AMPure XP, but it's unclear if they used the right bead ratio to minimise carry-through of the adapter dimers (I suspect not). We are doing some repeat work and also looking at another FFPE kit, so if I find out anything useful, will post it here at some point.
/Hedley
HedleyC is offline   Reply With Quote
Old 07-22-2014, 05:28 AM   #19
ServiceXS
Junior Member
 
Location: Leiden, the Netherlands

Join Date: May 2014
Posts: 6
Default

Hi,
Usually the yield of FFPE derived RNA is limiting.
How much FFPE-RNA did you start with for the Ribo Zero and how much was left over afterwards for the Truseq RNAseq sample preparation?
ServiceXS is offline   Reply With Quote
Old 07-24-2014, 08:49 PM   #20
woodydon
Member
 
Location: https://www.researchgate.net/profile/Woody_Lin

Join Date: Jan 2010
Posts: 52
Question

Dear all,

If you use RiboZero, will non-polyA RNAs dominate the RNA-Seq library? My friend told me that non-polyA RNAs are highly abundant. Will RiboZero end up dilute the polyA RNAs, which should be more important most of the time?

Bests,
Woody
woodydon is offline   Reply With Quote
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off




All times are GMT -8. The time now is 09:34 PM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2020, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO