XL+ vs. FLX FYI

[Anthony.287]

Just an FYI for anyone still using the XL+. It's a bad idea to run any FLX samples on the XL+ chemistry. From what was sort of explained to me by a 454 rep, the emPCR master mix and thermocycling conditions are optimized for one or the other, and cause problems when the protocols are mixed. I had some issues running PTPs of mixed samples with XL+ kits (some regions with 16s amplicons and some with Rapid libraries) where the amplicon data was quite poor and unexpected. When the same libraries were repeated with FLX chemistry and emPCR, the results were great. Not sure exactly what the cause(s) were, but I know it caused some bioinformatic headaches for our customers.

Best of luck! I'll see you on the Illumina forums!
Anthony

[martin2]

Hi Antony,
the amplicon vs. shotgun kits use different emPCR and hence sequencing primers. How was this ever supposed to work? You hust can't mix amplicon with shotgun samples (emPCR kit I vs kits II or III). What am I missing?
Martin

[Anthony.287]

That depends on the primers used to create the amplicons. You can use Lib-L primer sequences to create amplicons, which is what all of my previous 16s customers were doing. You end up with uni-directional reads but twice the coverage of Lib-A. This is the same kit used for rapid, PE and cDNA libraries.

I would never mix different customer's samples in the same emPCR or regions, but in different regions on a PTP.

What do you mean by "emPCR kit I vs kits II or III"? Are those Ion Torrent terms?

[martin2]

That depends on the primers used to create the amplicons. You can use Lib-L primer sequences to create amplicons ...

What do you mean by "emPCR kit I vs kits II or III"? Are those Ion Torrent terms?

Eh, I forgot that one can use the Lib-L approach if one uses the relevant fusion-primers, thank you for reminding me. ;-)

The emPCR kit II is a Roche term and stands for "amplicon A or paired-end" while emPCR kit III is for "amplicon B".

Back to your original post, I think it is rather obvious that longer PCR products are harder to obtain and are more prone to create offending secondary structures, dampening the emPCR or even sequenase. So, I find it rather obvious that some "to be amplified" regions could be failing under new protocol. In shotgun datasets this is hiding in the crowd of other, successfully amplified/sequenced random sequences.