problem with mRNA multiplex libraries

[sc10021]

I have been trying to generate mRNA-SEQ multiplex libraries using the multiplex kit from Illumina but my end repair, dA-tailing and ligation reagents are from NEB labs. I have no problem generating a smear post PCR if I use the standard Illumina PE reagents but do not see any PCR smears using the multiplex kit. I cannot see what is the difference in the protocol except using different adapters and primers. Can anyone help please? Thanks

[gford]

re: problem with mRNA multiplex libraries

I've just finished making 300 bp indexed paired end mRNA-seq libraries using these reagents and saw the expected smears on a Bioanalyzer, though they haven't been sequenced yet.

[josdegraaf]

How many pcr rounds did you both use?

[gford]

I titrated the number of Phusion PCR cycles to 17 for one sample, and 19 on the remaining 4.

[JMorlan]

We recently completed an RNA-seq study combining Illumina's RNA-Seq protocol and multiplex kit and running on the GAIIx. Our data quality was excellent in terms of total read yield, %Q30, % unique mapping, etc. The only caveat is that library yields are rather dramatically reduced (a few were right at 10 nM post-PCR) compared to the standard protocol but we always had more than enough to cluster.

Illumina has confirmed with us that the multiplex kit reduces library yields (hence the increase to 18 cycles in the protocol).

I would recommend that you quantitate rather precisely (qPCR or PicoGreen) as you may just have low-yield libraries.

Incidentally, we also use a "Home-Brew" version of Illumina's RNA-Seq kit, with reagents/oligos purchased from 3rd-party suppliers (the Illumina kits are a bit expensive) so I don't think it's a problem that you are using NEB kits.

-John

[josdegraaf]

We also did this several times now, using NEB as well as Illumina reagents and it works fine, we also had low yields but more then enough.

[amango]

problems with indexed paired-end mRNA-seq

JMorlan, gford, josdegraaf and anyone else,
I am working on a 'homemade' paired-end indexed mRNA-seq protocol. I am having trouble amplifying my adapter-ligated library (or maybe I'm having trouble with ligation). Would you be willing to send me your protocol and adapter & PCR sequences?

I am using the following adapter sequences, derived from another SEQanswers post (http://seqanswers.com/forums/showpos...5&postcount=13). 6-bp index tags are denoted by NNNNNN.

PEMx1 5’-ACACTCTTTCCCTACACGACGCTCTTCCGATCTNNNNNN*T
PEMx2 5’-pnnnnnnAGATCGGAAGAGCGGTTCAGCAGGAATGCCGAG

I plan to ligate different unique-index adapter set to my different libraries. And then enrich all libraries with these universal primers:

PCR PE1 5’-CAAGCAGAAGACGGCATACGAGATCGGTCTCGGCATTCCTGCTGAAC CGCTCTTCCGATCT
PCR PE2 5' - AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGA CGCTCTTCCGATCT

Am I correct that I do not need any PCR primers specific to my indexed adapters?

Any help would be appreciated!

[JMorlan]

Hi amango,

I should clarify that we did not use our 'Homebrew' RNA-Seq protocol on the multiplexing study I mentioned in my earlier post. For multiplexing we used Illumina's now obsolete Multiplexing Sample Preparation Oligonucleotide Kit. The protocol can be found on Illumina's website.

From your sequences it looks as if you are trying to have the barcode incorporated into the read itself. Just be aware that Illumina does not recommend this strategy (although that does not mean it can't be done).

To answer your question, since you are just adding a barcode to the end of your adapters, I would imagine you should be able to use the standard Illumina primers.

As for the problems you are having, some questions for you:

-Are you annealing your adapter oligos together (for example, with heating, slow-cooling) before you use them in ligation?
-Have you estimated the concentration of annealed adapter to use in ligation relative to your input template?

FYI, Illumina's new TruSeq RNA-Seq kits have multiplexing capability built in and are much less expensive than their earlier RNA-Seq Kits on a per sample basis (about $66/sample by my figuring). You might want to give that a try if you are still hitting a wall with your homebrew.

Good Luck,

JMorlan

[jowenMIT]

I agree with Jmorlan about properly annealing the adapters. I forgot to do this with some homemade adapters, and it resulted in an almost complete lack of proper ligation.