Truseq kit on PCR product

[Dun]

Hi,

Anybody has experience of using Truseq Kit (original Truseq DNA kit, not the TruSeq Custom Amplicon kit) on PCR product around 400~500bp?

It will be done without fragmentation. My question is that will the end-filling and ligation works? thx

[kcchan]

The TruSeq DNA kits work just fine on PCR products. The only minor issue is that you will not have strand specificity on your reads, so you'll have to take that into account with your analyses.

[Dun]

Or maybe I can just use gene-specific primers with Nextera sequence overhang and then do an additional PCR to add the index and sequencing adaptor.

Which one do you think is better?

Thank you.

[barrmur]

Or maybe I can just use gene-specific primers with Nextera sequence overhang and then do an additional PCR to add the index and sequencing adaptor.

Which one do you think is better?

Thank you.

If you are thinking about gene specific pcr I would include the illumina adapters and barcodes in the primers so you olmy need one pcr to create the libraries must like meta genomic preps. You will need a different reverse primer for every sample you wish to multiplex.

[Genohub]

The only trouble with the single PCR approach is that the primers you have to make are pretty long (includes the primer specific regions, flow cell binding, sequencing read primer binding sites, index and some regions designed to avoid dimer formation).

Do you know if the randomized Ns listed in the two PCR approach are still required with the latest Miseq software release?

- Genohub

[barrmur]

True the primers are pretty long (I think about £20 each) and can be expensive but I guess you need to weigh up the benefits of a one step library prep over the longer protocol. If you have not seen it already have a look at the protocol here,

http://www.earthmicrobiome.org/emp-standard-protocols/16s/

It gives designs for primers that include FC binding adapters, primer pads and index combinations. You did mention earlier that you were looking at a nextera style prep, presumably because of the need for extra indexes but the indexes here (12bp) allow for more diversity than the standard Truseq ones. Of course you could also modify these for use with dual indexes.

[microgirl123]

"Do you know if the randomized Ns listed in the two PCR approach are still required with the latest Miseq software release?"

@Genohub - I've heard that they're not necessary. However, I recently performed a run with the variable N primers, and still saw a lot of ups and downs with the intensities that lead me to believe the run quality would have been worse without the variable Ns.

[Dun]

Caporaso et al did a good job in explaining 16S amplicon assay. But for the general use, the two steps PCR might be more versatile since those long custom-designed primers are only good for the particular experiment. We need to make a whole new batch of primers for other amplicons.

Yet, from our experience,the issue of primer dimer in two step PCR is far more worse than one step.

[Genohub]

@microgirl, how much PhiX did you spike in when you used the variable N primers? Were the variable intensities at the beginning of the read?

[microgirl123]

@ Dun - we haven't had a problem yet with primer dimer in the 2 step process. I encourage people with problems with primer dimer in the 1st step to use an 0.6x Ampure cleanup because the primer dimers are too big to be removed with a spin column or regular Ampure concentrations.

@ Genhub - I used a 5% phiX spike. The BaseSpace charts are attached (they're small, I'm having trouble with uploading!). The variable intensities are throughout the entire read - I would have expected this without the variable Ns.

[Genohub]

amplicon intensity and PhiX spike in

Yes, I see the intensity spikes. I wonder if others are seeing the same thing with amplicons using a 5% PhiX spike. Are you considering increasing your spike on the next run? I've spiked in as much as 10-15% PhiX with 16S amplicons using the updated MiSeq software.

- Genohub

[microgirl123]

I think I'll probably leave the 5% phiX spike as is, the Q30 was 94% so it doesn't look like it was a problem. I was just surprised to see the spikes with the offset primers.

[Dun]

So you added variable N in the reverse primers and did 5% phiX spike in.

I have a question about the added N. How many N does it need?
For example: I have a single amplicon, 20 different samples, should I design four different reverse primers with no N, 1N, 2N and 3N respectively and spread them across 20 samples?

Thx microgirl123

I think I'll probably leave the 5% phiX spike as is, the Q30 was 94% so it doesn't look like it was a problem. I was just surprised to see the spikes with the offset primers.

[microgirl123]

No. You should mix all your forward primers together by equal volumes (no N, 1N, 2N, and 3N) into a "forward primer cocktail" and all your reverse primers together into a "reverse primer cocktail." Then use those primer cocktails to perform one PCR on each sample.

So each sample will be offset by a variable number of bases.

[Dun]

Get it. I thought putting Ns in reverse primer is good enough though.

No. You should mix all your forward primers together by equal volumes (no N, 1N, 2N, and 3N) into a "forward primer cocktail" and all your reverse primers together into a "reverse primer cocktail." Then use those primer cocktails to perform one PCR on each sample.

So each sample will be offset by a variable number of bases.