real time PCR primer designing problems,again - is NCBI's primer blast believable enough？ (Apr/23/2009 )

I once asked sth. about the primer desining for a certain gene which containing 13 isoforms several months ago, and sb. told me to use the NCBI blast tool(http://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE_TYPE=BlastSearch&PROG_DEF=blastn&BLAST_PROG_DEF=megaBlast&BLAST_SPEC=blast2seq)
that's an excellent tool which enabled me to got what i wanted.

however, today the senior in my lab told me that because the isoforms 3and 4 are too homologous that even the pair of the primer i designed for isoform3 can only find one BLAT product in the genome, it is possibe to bind to the gene of isoform4 when doing PCR,because the reverse primer fits the isoform4 quite well.

my question is : is the senior right? should i redesign for isoform3?

First I would use the blast2seq (the link you posted) to align both isoforms to see where is the holomogy and design the primers in nonhomologous regions. (or align all the isoforms in ClustalW to see a unigue sequences among them)

Then use blastn to check the primers to see if they are really specific.

Generaly when designing in genes with homology"
- it should be enough if one of the primers doesn't bind the homologue, because you need a pair to amplify a region, but you maybe need to put more of that primer to the reaction
- if you can't find larger nonhomologous part, design primers to have one or more base mismatch on the 3' end, that's important for the elongation
To be sure it's better to have both primers with at least 3' mismatch.

If the isoforms are in the database, blast should show them. I would trust blast.

bear on Apr 23 2009, 12:54 AM said:

i see, thank you very much!!