The resolution of your gel is not sufficient to make out the mol wt of the 2 bands, but the incorrect band looks like 200bp - double that of the correct one. Is this DNA or cDNA that is being amplified? what are you amplifying ? Have you BLASTed your primers to see if they have alternative binding sites?

I presume you have tried different (raised) annealing temperatures etc . Does the melting profile indicate 2 very different products ?

You could have 1) mispriming 2) alternative priming 3) alternative splicing (cDNA) 4) some form of dimerization

Ian

thanks for your answers. i amplified cDNA and amplifying is over than 2. and image of melt is:

In your case, the primers are not specific to your target. If you are using the cDNA as a template, you better make the primers as follow;

1. select primers mainly on exon-exon junction (so that no genomic DNA can amplify)

2. make the product length between 100 to 200 bp for real time PCR (can also use for sequencing)

3. check the primer specificity by using PrimerBLAST. It must be 100% specific with your gene and no on the other gene. In case if it is not possible, select not specific for other gene especially at 3' end.

4. If you have a positive control cDNA, use it to identify template or primer problem. Sometime, your template may be not good during total RNA to cDNA depending on the kit.

5. check it on the gel electrophoresis and sequencing for specificity

Good Luck !

thanks.

i Blasted my primer in NCBI and it is 100% specific to my target gene. in BLAST it is not have non specific product. i attached primer BLAST in Word format.

Well, I just did a blastN (similar sequence) of the primer sequences based on the Word document. While your primers do match 100% identity to your intended target (TGF-B2 variant 2, I assume), the E-value is rather high at 0.009 (forward) and 0.091(reverse).

it means that the the annealing region for your primers is not very specific for your gene, and other genes might have a similar region as well (e.g. 90% identity). Remember that PCR is not just based on specificity, but based on the abundance of transcript. If a 90% identical gene is 100x more abundant than a 100% identical gene, amplification will favour the 90% identical gene

Also note that the blast hits reports transcript variants, so it is possible that other variants of different length could be amplified.

My suggestion is to increase the annealing temperature to the highest possible (don't assume based on the spec sheet, run a gradient temperature PCR). Or design new primers that have lower E-value while maintaining high identity.

Thanks mr.lee

i want to do your advice and report it for u

I often use Roche Universal Probe Library Design Software available at

http://www.roche-applied-science.com/webapp/wcs/stores/servlet/CategoryDisplay?catalogId=10001&tab=&identifier=Universal+Probe+Library&langId=-1#tab-3

to design my primers. Just have a look at Roche's suggested temp profile for qRT-PCR if you are not using it already.

have you or anybody in your lab done this reaction before? How sure are you that the annealing temp that you are using is indeed correct. If this is the first time, you are doing this reaction and your annealing temperature has been determined mathematically, then like others suggested, you need to set up a gradient to determine your annealing temperature.

If you have experimentally determined the correct annealing temperature, you could use a positive control to show that the you get only one band at the said temperature and something is wrong with this reaction. It would also help if you could get a calibration report for your instrument to know if everything is fine.