How to multiplex Illumina matepair libraries?

versa · 10-20-2011, 08:39 AM

Hi,

Could anybody help me with this: We would like to make a multiplexed matepair library (with insert size of 3kb). We were trying to do this by using the multiplex sample prep kit from Illumina (since there is no TruSeq protocol yet for matepair sequencing) but we fail to get good results. Has anybody ever done this? Or is there something else we can do instead of this?

Thanks!

pmiguel · 10-20-2011, 09:29 AM

We used a Roche 454 "paired end" (really mate end) preparation kit but used TruSeq adapters instead of Roche adapters. (And, obviously, sheared the circularized DNA to a smaller size than called for the the 454 protocol.)

It worked okay. Actually the data it produced was very good, but the "insert" sizes were much shorter, for the most part, than the size of the initial DNA fragmentation. (1.4 kb instead of 7 kb). But we did not do a size selection prior to circularization, so we will try that next time.

--
Phillip

versa · 10-24-2011, 12:00 AM

Hi Philip,

Thanks for your reply! If possible, could you give me some more details about your customized protocol?

Thanks!

pmiguel · 10-24-2011, 06:01 AM

Well, there is not much more to it than what I described.

Actually I heard that a core at another university was doing this. We had a Roche PE kit on hand and pretty much followed the directions there. Obviously, during the second shearing step (post-circularization) we sheared the DNA smaller than called for in the Roche protocol. Oh, do I need to add, that we do not add the extra DNA from an unknown source for the second shearing that is called for in the Roche protocol? We used the sonication protocol we normally use for our Illumina PE libraries. Then we had to add A-tailing to the "end polishing" part of the protocol and used Illumina TruSeq adapters.

Because the number of read pairs we would collect with an Illumina sequencer was 10-100x more than with a 454, we pooled a couple of reactions. Nevertheless it looked like we were bottoming out the resulting library with only 1/2 of a lane of sequence. Also, like I said earlier, instead of having the pairs mostly 7 kb apart, they were mostly 1-2 kb apart. Suggesting that we needed to add an agarose gel size selection to the protocol. And, as is apparently not uncommon for libraries of this type, about 1/3rd of them ended up PE, instead of ME.

I don't like the Illumina method because they don't use a center adapter that allows identification of the circularization ligation junction informatically. But their method may be more cost-effective, I suppose. I have not run the numbers.

--
Phillip

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Multiplex adapters and indexes for Illumina HiSeq	arg	Sample Prep / Library Generation	18	10-07-2013 03:53 AM
MatePair library	Lisamae19	Illumina/Solexa	1	09-26-2011 10:19 AM
comparison of Illumina libraries	PFS	Bioinformatics	1	03-17-2011 09:46 AM
problem with mRNA multiplex libraries	sc10021	Illumina/Solexa	8	01-27-2011 02:30 PM
searching 454 (single read) + Illumina paired-end or matepair data	boetsie	General	0	10-04-2010 02:45 AM

10-20-2011, 08:39 AM	#1
versa Junior Member Location: Europe Join Date: Feb 2011 Posts: 4	How to multiplex Illumina matepair libraries? Hi, Could anybody help me with this: We would like to make a multiplexed matepair library (with insert size of 3kb). We were trying to do this by using the multiplex sample prep kit from Illumina (since there is no TruSeq protocol yet for matepair sequencing) but we fail to get good results. Has anybody ever done this? Or is there something else we can do instead of this? Thanks!

10-20-2011, 09:29 AM	#2
pmiguel Senior Member Location: Purdue University, West Lafayette, Indiana Join Date: Aug 2008 Posts: 2,304	We used a Roche 454 "paired end" (really mate end) preparation kit but used TruSeq adapters instead of Roche adapters. (And, obviously, sheared the circularized DNA to a smaller size than called for the the 454 protocol.) It worked okay. Actually the data it produced was very good, but the "insert" sizes were much shorter, for the most part, than the size of the initial DNA fragmentation. (1.4 kb instead of 7 kb). But we did not do a size selection prior to circularization, so we will try that next time. -- Phillip

10-24-2011, 12:00 AM	#3
versa Junior Member Location: Europe Join Date: Feb 2011 Posts: 4	Hi Philip, Thanks for your reply! If possible, could you give me some more details about your customized protocol? Thanks!

10-24-2011, 06:01 AM	#4
pmiguel Senior Member Location: Purdue University, West Lafayette, Indiana Join Date: Aug 2008 Posts: 2,304	Well, there is not much more to it than what I described. Actually I heard that a core at another university was doing this. We had a Roche PE kit on hand and pretty much followed the directions there. Obviously, during the second shearing step (post-circularization) we sheared the DNA smaller than called for in the Roche protocol. Oh, do I need to add, that we do not add the extra DNA from an unknown source for the second shearing that is called for in the Roche protocol? We used the sonication protocol we normally use for our Illumina PE libraries. Then we had to add A-tailing to the "end polishing" part of the protocol and used Illumina TruSeq adapters. Because the number of read pairs we would collect with an Illumina sequencer was 10-100x more than with a 454, we pooled a couple of reactions. Nevertheless it looked like we were bottoming out the resulting library with only 1/2 of a lane of sequence. Also, like I said earlier, instead of having the pairs mostly 7 kb apart, they were mostly 1-2 kb apart. Suggesting that we needed to add an agarose gel size selection to the protocol. And, as is apparently not uncommon for libraries of this type, about 1/3rd of them ended up PE, instead of ME. I don't like the Illumina method because they don't use a center adapter that allows identification of the circularization ligation junction informatically. But their method may be more cost-effective, I suppose. I have not run the numbers. -- Phillip