emPCR thermal cycler settings

[suefo]

We are experiencing high enrichment with our emPCR as well as a high %Short when we achieve an enrichment that allows us to sequence (GS Junior). We have examined our PCR plates, we have used different thermal cyclers and we have tried different tissuelysers / turraxes (looking at emulsion formation) and nothing seems to provide any clues. The only parameter we haven't really played with are the temperatures, times and number of cycles the thermal cyclers of the program.

We are using the temps, times and cycles as outlined in the Roche protocol for em PCR (LVs, MVs, SVs and Junior kits). We are looking at 16S Amplicons. Has anyone else looked at this and use different temps, times and cycles?

[RCJK]

Hi suefo, you should only need to adjust the emPCR parameters if you are working with short or long amplicons. Have a look at their amplicon guidelines as well as technical bulletin 2011-001 (geared towards Lib-A amplicons though). I had a project in which the amplicons were ~180bp-300bp and results improved a lot when we used the short emPCR recipes and cycling parameters compared to the "normal" conditions and recipes.

[suefo]

RCJK... thanks. I will check out the technical bulletin. We are looking at amplicons in the 550-600 bp range and we are using the Lib-L kits.

[RCJK]

Also, how high of a %short are you seeing? I do tend to see many more reads filtered out for amplicon samples than shotgun samples. Are the amplicons made with fusion primers or ligated adaptors? It may be worth doing a 2nd round of small fragment removal with the Ampure XP beads. If enrichment recoveries are high, reduce the cpb input.
I've sequenced Lib-L amplicons of ~550bp fine using the normal emPCR conditions, though that is approaching their suggested cut off for using the long emPCR conditions. I have been told though that they have not been tested for Lib-L amplicons. Maybe you could follow the Lib-L emPCR conditions of the FLX+ protocols which should be for longer fragments (but maybe yours are not long enough).
Lastly, if the run yields are low, depending on your application you can modify the filtering parameters to trim the amplicon reads instead of rejecting them. This should also be addressed in the updated amplicon guidelines.
Hope this helps!

[suefo]

Thanks again RCJK.
Our %Short is around 60% (%Dot+Mix is at 15%).
We do perform an initial AMPure, run the products on the Agilent, pool DNA, re-purify a second time using AMPure and then run a final high-sensitivity Agilent test. Our samples appear to be extremely clean.
We have lowered our cpb down 0.05 and our %enrichment is between 13-18%. This is of course using the Junior and the Junior kits.
We also perform a second run of analysis loosening the filtering parameters which really only impacts the %short reducing it 4-8% from the original Amplicon analysis.
I believe we are using ligated adaptors - but I will double check on that.

Roche has wanted us to try their FLX+ kits. Perhaps I should suggest they send us a few test samples to see if that helps.

I'm new to this technology. My understanding is that once the sequencing is completed they run it through another analysis package that trims the bases down even further and that there may not be too much concern about the %short. We are currently looking at sequencing data obtained off our Junior with previous data generated off the FLX to see if this is the case.

[suefo]

Ooops... I stand corrected on our length. I just re-checked the agilent data and it appears the DNA being submitted for testing is 650 bp (that's including the PCR adaptors, barcodes and keys).

[flxlex]

Roche has wanted us to try their FLX+ kits. Perhaps I should suggest they send us a few test samples to see if that helps.

Are you sure? AFAIK, FLX+ is not available for either GS Junior, or amplicon runs?

[[email protected]]

60% reads is very higher. The Roche protocol shows more than 20% is failed.
Do you try to change primer volume? or dilute lower concentration of DNA library?