Paired end Illumina prep problems?

[James]

Hi people.

I was wondering if anybody had had any problems with the illumina protocol for making the libraries for paired end sequencing. We previously did some ChIP-seq before (not paired end) and the first time we ran into problems with primer dimers and had to lower primer and adapter concentrations.

This time we are sequencing DNA (not ChIPed), are there any similar common problems? And do you do the PCR enrichment, I hear some people are missing this out and getting good results?

Cheers, James

[SeqR&D]

Hey James,

What is PCR enrichment? I have made many PE libraries and do not run into too many issues. I follow the illumina protocol. Have you tried making these libraries yet? How much DNA are you starting with?

[James]

Hi,

By PCR enrichment. I mean one of the last steps in the protocol where you do 10-12 PCR cycles. I just read the protocol in a bit more detail it seems it is actually essential for adding sequence on the end of the adaptors. Never mind.

I'm aiming to start the library prep with 5ug. Do you do the 10 cycles the protocol recommends?

Thanks, Just wanted to scope out see If there was anything in the protocol that everyone changes before starting. I haven't made these libraries yet.

Cheers, J

[SeqR&D]

Okay...I get it.

Yes, I do this PCR...both to add the correct sequence for sequencing, and to ensure that you have library with correctly orientated adapters. I generally do 14 cycles, but this really depends on you. I have done up to 16 cycles. (I may have different applications though) The lower the better, I would imagine...less amplification = better diversity. Also, I have started using Ampure beads instead of column purifications...should give you more yield.

Good luck!

[cliffbeall]

PCR free method

A method without the PCR step was published by Kozarewa et al in Nature Methods 6, 291 - 295 (2009). Link The different ends are inherent in the "Y" ligation, the solution PCR is only needed to get enough material to quantify before running. The authors get around that by quantitating using Q-PCR. It does require custom adapters that are longer than the commercial ones.

I was thinking of trying it- I am trying to get as accurate as possible reflection of the input. Its probably not a routine kind of thing one would want to do, unless there is some kind of specific need.