Protocol Online logo
Top : Forum Archives: : siRNA, microRNA and RNAi

Problem with design of shRNA - (Jun/10/2005 )


I'm quite new to the field of RNAi. I have used the program on the ambion website to create Target sequences. Afterwards I wanted to check for matches within the genomic DNA/RNA via the Blast Link on the Ambion page to find out, which target sequences I should use. The guidelines on the Ambion site say, that you have to have at least 4 mismatches with any other gene within the organism you want to use the shRNA in. I found sequences with maybe 7 to 8 mismathces to my target sequence but unfortunatelly the mismatches are always at the start or the end of the sequence. so there are stilll 14 or 15 bases matching in a row.

My question now is if this is a problem or not that there are still long matching sequences because I have no idea.

Thanks for your help!



I don't know the answer to your question, but this website has a lot of shRNA adivse that could help you with an answer:


Hi Stardust,

I understand your concern. If the mismatches are only at the 3' end of the guide strand, while 5' end has perfect match to unrelated sequence, that could be of problematic. For example, we have this gene


red is a imaginary siRNA target

We will have this double-stranded siRNA


The blue strand is the GUIDE strand which guide the RISC complex to its target.

If the 5' portion of the guide strand has perfect match to other sequence it may initiate RNAi. So in your case, if the mismatches are at the 3' end of the target sequence (5' end of the guide strand), the siRNA target is OK; otherwise, unaceptable.

The above is based on published literature and also on my own observation in which I mutated 5 base pairs at the 5' portion of a siRNA relative to the guide strand, the mutation totally disrupted the RNAi effect. I also mutated 5' bps at the 3' portion of my siRNA relative to the guide strand, this mutation didn't affect the siRNA's action at all.

Hope that helps and is clear to you.


i agree with pcr man. Ive read in publications that at least 13 successive nt should be without mismatch and 4-5 mismatches are allowed in the 3' part of the siRNA sequence...


thanks a lot for the answers. In my sequences I have mismatches at 3' and 5' end put they fit for 13 or 14 bases in the middle. So if I understand you correctly, that should not be a problem.