RNA-seq: rRNA Depletion, Double riboZero or Double RiboMinus

cvhove · 06-25-2012, 11:47 PM

Hi,
I did a RNA-seq sample on Illumina GA2x: 15 years old Tumor FFPE sample. The RNA extraction give a very low RIN score analysing on the BioAnalyser RIN +/-1.5.
I used the EPICENTRE combination of RIBO-ZERO with SCRIPTSEQ v2. Good results 30-50% mapped reads, but still 35% related to rRNA. Is there a protocol to remove more rRNA out of the sample. Doing the Depletion twice, combinations Ribo-Zero with RiboMinus or other manors to increase the mapped reads?
thanks
c

JakobHedegaard · 06-27-2012, 01:13 AM

Hi cvhove,
We have also done RNA-Seq on old (up to 20 years) tumor FFPE samples using Ribo-Zero in combination with ScriptSeq. In general we see 70-90% of the reads mapping to the human transcriptome set from ensembl (not including rRNA) and a variable fraction of the un-mapped reads mapping to rRNA after de-novo assembly.
Be aware that the rRNA reads can be due to 1) insufficient removal of rRNA during Ribo-Zero, or 2) "leaking" of the Ribo-Zero probes into ScriptSeq resulting in rRNA reads. Try to contact Epicentre for support.
Cheers, Jakob

ZWB · 06-27-2012, 09:22 AM

Hello,

We have had very good luck removing rRNA from degraded samples using DSN treatment. In degraded samples we find that DSN works much better than Ribo-Zero often dropping the final rRNA to <5%. I would strongly recommend against multiple rounds of Ribo-Zero as we have lost entire samples attempting it.

JakobHedegaard · 06-28-2012, 12:04 AM

Hi zwb,
We were also considering DSN, but skipped it due to the potentiel effects on higly expressed transcripts. Have you tried to compare profiles of DSN vs non-DSN to see if the expression levels of highly expressed transcripts are underestimated in DSN?
/Jakob

ZWB · 06-28-2012, 09:57 AM

We find that the most highly expressed genes are still the most highly expressed genes after DSN treamtent. So if you look at a metric like the top 10 most highly expressed genes before and after DSN the lists contain the same genes in the same order. While the expression levels of the most highly expressed genes do even out somewhat, it seems like the vast majority of what you are removing is rRNA and not the mRNA you are interested in. In addition, we find many hundreds of low expression level genes that we were not able to see before regardless of sequencing depth. I hope this helps.

ZWB

ehdezsanabria · 08-22-2014, 04:18 AM

Hi Jakob,
We used the Turbo DNase from Ambion and followed by clean-up with RNAeasy mini kit. Then we used the Terminator exonuclease (because we do not have access to the RiboZero kit) and constructed the libraries with the ScriptSeq kit. Of course our results were not good and the libraries were degraded. Any advice on how to deplete the RNA without using Ribozero? I am planning to remove the RNA with the on-column protocol from the RNAeasy mini kit from Qiagen. However, I am not sure if I can directly proceed to the library construction without depleting. Thanks a lot for your advice!!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
RNA-seq libraries - double peak after size selection	MLog	Sample Prep / Library Generation	23	03-08-2012 10:22 AM
Double band in ChIP-seq lib prep for Illumina	zhaoj	Sample Prep / Library Generation	8	08-01-2011 10:04 AM
RNA-Seq: RNA-Seq Analysis of Gene Expression and Alternative Splicing by Double-Rando	Newsbot!	Literature Watch	0	03-03-2011 02:00 AM
RNA-Seq: Multimodal RNA-seq using single-strand, double-strand, and CircLigase-based	Newsbot!	Literature Watch	0	12-24-2010 02:13 AM
ChIP-Seq: Efficient Double Fragmentation ChIP-seq Provides Nucleotide Resolution Prot	Newsbot!	Literature Watch	0	12-15-2010 02:00 AM

06-25-2012, 11:47 PM	#1
cvhove Junior Member Location: Belgium Join Date: May 2011 Posts: 3	RNA-seq: rRNA Depletion, Double riboZero or Double RiboMinus Hi, I did a RNA-seq sample on Illumina GA2x: 15 years old Tumor FFPE sample. The RNA extraction give a very low RIN score analysing on the BioAnalyser RIN +/-1.5. I used the EPICENTRE combination of RIBO-ZERO with SCRIPTSEQ v2. Good results 30-50% mapped reads, but still 35% related to rRNA. Is there a protocol to remove more rRNA out of the sample. Doing the Depletion twice, combinations Ribo-Zero with RiboMinus or other manors to increase the mapped reads? thanks c

06-27-2012, 01:13 AM	#2
JakobHedegaard Member Location: Aarhus, Denmark Join Date: Mar 2008 Posts: 57	Hi cvhove, We have also done RNA-Seq on old (up to 20 years) tumor FFPE samples using Ribo-Zero in combination with ScriptSeq. In general we see 70-90% of the reads mapping to the human transcriptome set from ensembl (not including rRNA) and a variable fraction of the un-mapped reads mapping to rRNA after de-novo assembly. Be aware that the rRNA reads can be due to 1) insufficient removal of rRNA during Ribo-Zero, or 2) "leaking" of the Ribo-Zero probes into ScriptSeq resulting in rRNA reads. Try to contact Epicentre for support. Cheers, Jakob

06-27-2012, 09:22 AM	#3
ZWB Member Location: Livermore, CA Join Date: Apr 2011 Posts: 27	Hello, We have had very good luck removing rRNA from degraded samples using DSN treatment. In degraded samples we find that DSN works much better than Ribo-Zero often dropping the final rRNA to <5%. I would strongly recommend against multiple rounds of Ribo-Zero as we have lost entire samples attempting it.

06-28-2012, 12:04 AM	#4
JakobHedegaard Member Location: Aarhus, Denmark Join Date: Mar 2008 Posts: 57	Hi zwb, We were also considering DSN, but skipped it due to the potentiel effects on higly expressed transcripts. Have you tried to compare profiles of DSN vs non-DSN to see if the expression levels of highly expressed transcripts are underestimated in DSN? /Jakob

06-28-2012, 09:57 AM	#5
ZWB Member Location: Livermore, CA Join Date: Apr 2011 Posts: 27	We find that the most highly expressed genes are still the most highly expressed genes after DSN treamtent. So if you look at a metric like the top 10 most highly expressed genes before and after DSN the lists contain the same genes in the same order. While the expression levels of the most highly expressed genes do even out somewhat, it seems like the vast majority of what you are removing is rRNA and not the mRNA you are interested in. In addition, we find many hundreds of low expression level genes that we were not able to see before regardless of sequencing depth. I hope this helps. ZWB

08-22-2014, 04:18 AM	#6
ehdezsanabria Junior Member Location: belgium Join Date: Aug 2014 Posts: 3	Hi Jakob, We used the Turbo DNase from Ambion and followed by clean-up with RNAeasy mini kit. Then we used the Terminator exonuclease (because we do not have access to the RiboZero kit) and constructed the libraries with the ScriptSeq kit. Of course our results were not good and the libraries were degraded. Any advice on how to deplete the RNA without using Ribozero? I am planning to remove the RNA with the on-column protocol from the RNAeasy mini kit from Qiagen. However, I am not sure if I can directly proceed to the library construction without depleting. Thanks a lot for your advice!!