SEQanswers

Go Back   SEQanswers > Sequencing Technologies/Companies > Illumina/Solexa



Similar Threads
Thread Thread Starter Forum Replies Last Post
Gel-size selection in NimbleGen exon capture mitchelS Illumina/Solexa 3 10-26-2012 08:50 AM
Post PCR Enrichment Gel Selection for ChIP-Seq Daytwa Sample Prep / Library Generation 7 05-10-2012 07:06 AM
Nextera Problem: Odd fragment size after gel selection Kitty Sample Prep / Library Generation 3 09-15-2011 03:54 AM
Size Select Protocol without using Gel zeineen Sample Prep / Library Generation 1 03-27-2011 11:16 AM
Need help: Weird size selection chromatin Sample Prep / Library Generation 0 11-15-2010 03:24 AM

Reply
 
Thread Tools
Old 11-28-2011, 12:39 AM   #1
puggie
Member
 
Location: Europe

Join Date: Nov 2011
Posts: 52
Default PE (in gel) Size-Selection

Hello,

I'm new to this forum, however have used it as a guest for some time. In our lab we want to do paired end RNA-Seq on Illumina (which we have access to). I have made some libraries with the ScriptSeq mRNA protocol from Epicentre. Actually I have not enriched for mRNA, but removed rRNAs. So it will be whole-transcriptome.

The size distribution I get looks somewhat similar to the examples in the ScriptSeq protocol. That is I have a broad distribution from approx. 200-1500 bp (DNA high sensitivity chip on BioAnalyzer 2100).

Now I'm reading that it is best to size-select within a narrow range (let's say 350-450 bp) to get optimal bridge-amplification. If we consider that the fragmentation is unbiased (which it may not be), size-selection shouldn't be a concern for quantitative output from sequencing data or?

I'm not sure how to proceed. The plan is to "fragment" the library on agarose gel electrophoresis. So I can get distributions from e.g. 200-350, 350-450, 450-550 bp and so on. How do you guys do this? Or do you size-select at all? I believe that paired-end sequencing on RNA-seq for our pursoses would not make sense if some overlap between the reads is not present so they can be assembled in full ~2x150 bp.

Hope somebody can give me some inputs or considerations on this issue.
puggie is offline   Reply With Quote
Old 11-28-2011, 03:58 AM   #2
pmiguel
Senior Member
 
Location: Purdue University, West Lafayette, Indiana

Join Date: Aug 2008
Posts: 2,313
Default

What you describe is not technically "fragmentation", but "fractionation".
RNA's 2' hydroxyl groups makes it easy to chemically fragment it (nucleophilic attack of the 2' hydroxyl on the phosphate backbone causes strand breakage). So most RNAseq methods use heat+divalent cation to fragment the RNA into an amenable size range, followed by random prime 1st and 2nd strand synthesis. If you fragment aggressively enough then all that is needed is a method to remove constructs below your desired size range. This can be done with ampure, although gels also work.

"Random" in this case is not as random as one might wish -- presumably annealing characteristics of the hexamers causes some bias. So, if you have a limited number of samples, sonication can be used to fragment cDNA into a desirable size range.

--
Phillip
pmiguel is offline   Reply With Quote
Old 11-28-2011, 04:06 AM   #3
puggie
Member
 
Location: Europe

Join Date: Nov 2011
Posts: 52
Default

Phillip thx for you reply. Yes I meant fractionation on gel not fragmentation. Fragmentation I did following ScriptSeq protocol, 5 min at 85C, perhaps this needs optimization. Still do you guys size-select or finetune fragmentation (or both) to get within the desired size range. Actually I do not understand why this aspect is not covered in the ScriptSeq protocol.
puggie is offline   Reply With Quote
Old 11-28-2011, 04:24 AM   #4
pmiguel
Senior Member
 
Location: Purdue University, West Lafayette, Indiana

Join Date: Aug 2008
Posts: 2,313
Default

For Illumina we construct most of our libraries such that they are optimized for de novo transcriptome -- this means using a shorter incubation at elevated temp for fragmentation (4 minutes instead of 8 minutes). That is, larger cDNAs so that PE reads are optimally used. And, yes, we nearly always do a size selection. Generally we pool similar (but indexed) libraries in Pippin prep lanes. But sometimes we just use a normal agarose gel method.

Normally we do not take multiple fractions, however. Arguably this might bias the libraries against longer transcripts. But our standard method of ribo-depletion (polyA) already will cause massive biases of various sorts. I am doubting the size cut will cause issues detectable against systematic biases already introduced.

--
Phillip
pmiguel 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 06:32 PM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO