Question about RealTime PCR Primer Design - (Jan/11/2011 )

Hello out there,

I am having problems with designing primers for my studies.
Right at the moment I'm working with PerlPrimer but i got a little upset with my results.
I wanted to design very specific primers so I was looking for conservated gene loci in my gene ,which has several transcripts, and then I tried to build a primer which overlapses a Exon-Exon binding site by seven bases which was still sensitive for all my different transcript variants.

Then i blasted these primers with NCBI and there were some primers which were exactly specific just for my target transcript variants of my gene. Unfortunately they had very mean binding qualities (Hairpin, SelfDimer...) The other primers were also sensitive for my gene but they also bound to other genes with several base mistakes. Unfortunately, they had better binding qualities.

My question is: How tolerable shall I be with Blast results of non targeted genes, when they differ with 1,2,3,4,5,6,7 bases from my target sequence.

I would be grateful for any help,
bye Alex

If you leave the mismatches at the very 3' end of the primers, even one mismatch is sufficient for specificity.