Nested PCR amplicons

[AEscobar]

Our PCR protocol requires nested PCR in order to be more sensitive. We would like to amplicon sequence such nested PCR fragments but we were wondering how much 'bias' it would be introduced by the second round PCR? Would this favor amplification of major variants and 'elimination' of the minors just because of the frequency of each variant, thus reducing the complexity (number of haplotypes) after deep sequencing?
Has anyone tryed amplicon sequence nested amplicons?
Is it recommended by Roche?

[pmiguel]

Roche is extremely circumspect in these matters--they will not have officially offered an opinion one way or another.

In any case you have articulated a concern. Seems like a reasonable concern to me. You can test whether there is a problem using spike in experiments.

Or validate any results you see using another method afterwards.
--
Phillip

[AEscobar]

pmiguel

Thanks for your reply!

You can test whether there is a problem using spike in experiments.

Or validate any results you see using another method afterwards.

Would you mind providing more details about both suggestions?

[pmiguel]

Thanks for your reply!

Would you mind providing more details about both suggestions?

Sure. But I am more or less stating the obvious. Please don't expect any deep insights here.

In a spike in experiment you add known amounts of sample from a known source at at different levels of dilution to your assay. If your assay faithfully reflects those known levels then you have reason to suspect that it will also do so for the samples you wish to test.

A comprehensive spike in test would likely be difficult to design. So many instead go the validation route. In this route you trust your new, untested, assay only to provide candidates for assays that are known to be reliable. That is, you might use an RNA seq experiment to "scan" through the transcriptomes of your samples under conditions of interest. Then based on an arbitrary threshold (say the 10 transcripts showing the highest variation (fold change) under the conditions you are studying) you choose candidates for further testing. For example test their levels specifically using qPCR.

The advantage of this method is that you can use powerful new technologies for "descriptive" rather than analytical purposes. Then you don't need to validate the new technologies up front yet you can benefit from their power.

--
Phillip

[AEscobar]

Thanks!

We are already designing an experiment where we will be using some of our plasmids as controls.