I am multiplex PCR-ing 12 SNPs before sequencing on the MiSeq.  The primers were originally designed to be used on an Fluidigm EP1 system and have been adapted to include the universal overhangs from Illumina.

All SNPs amplify successfully, but appear to be fixed for the heterozygote genotype in all individuals, which is not correct (verified using the EP1).

During troubleshooting, I discovered that the STA/LSP primers designed by Fluidigm do not always correspond to F/R primers, respectively.  As such, in some SNPs, the Forward and Reverse primers have been switched around.

This, I assume, would result in PCR failure of the targeted sequence.  Can anyone confirm this is the case? 

However, when sequenced, the post-run script, which looks for and counts the number of times it finds SNP-specific probe sequences (going from the start of the Forward primer (which could be the reverse primer) to 2 bp after the SNP) in the individual's fastq file, appears to be able to find the target DNA in question.

Does anyone have an explanation for this, as it has absolutely baffled me.

Many thanks!