You should explain better what is your question, maybe provide some picture of what you want to do.

Primers cover such sequence as you design them. If you want to cover certain part of sequence, simply design homologous primers (not the whole sequence must be homologous).

Sounds like you want to do some sort of variant-discrimination based on SNPs to look at gene expression by designing primers so the mismatch is the 3' end of the primer.

It's technically possible, but doesn't mean that it will work.

Designing qPCR primers that have 95-105% efficiency that are specific to a region of interest is already very hard to get correct. Being restricted to a specific region makes it even harder. I'd design a few primer pairs & see if you can get the mismatch in the "forward" for some pairs & "reverse" for others. That will increase the likelihood that this could work.

Also, some 3' mismatches are actually well-tolerated by DNA polymerases, especially ones that proof-read. Not sure if that's true for the polymerase in qPCR kits, you'd have to do some investigating.

Might be easier to do a probe-based qPCR if the mismatches are destabilizing enough.

Good luck!

Thank you

The problem is that the nucleotide sequences that could distinguish our novel sequence from that of annotated database sequence are located at the beginning and end of our target sequence, and thus no suitable location for primers binding at these sites. We tried to use modified primers/ probe oligo such as locked nucleic acid ( LNA) to force primer bind these specific sites but didn't work well. I am searching for an alternative methods Katie A S Burnette Tomáš Hluska

Salem Baldi can you provide us the sequence or picture of the alignment to get some idea?

So what is your aim? To use RT-PCR to differentiate between RNA molecules?

Are you including any 5' or 3' UTR in your target sequence (assuming eukaryotic)? Those regions can be used in qPCR