Variant found in priming site

[HeidiJTP]

I sequenced barcoded PCR amplicons on Ion Torrent PGM, and a subset of samples were also Sanger sequenced to confirm variable sites.

I first noticed that a variant was showing up in the forward priming sequence for one locus in some of the Sanger data and thought it was just a sequencing artefact since the site is not variable in any of my Ion Torrent reads for those same samples.

However, I found the same issue in the reverse priming sequence for both Sanger and Ion Torrent reads at another locus. The variant does not occur in all of the Ion Torrent reads but it is at the same site as in the Sanger reads. Could this still be an artefact? Am I correct in thinking that a true variant could not be recovered when it occurs within the PCR primer binding site?

[Bukowski]

Am I correct in thinking that a true variant could not be recovered when it occurs within the PCR primer binding site?

Short answer yes, a well known problem with amplicon approaches. Although 'could not' might be better phrased as 'may not'. Depends where it is in the binding site. However I have certainly run across this issue.