Re: pcr_seq

[Martin Asser Hansen]

It's a tricky question both in vivo and in silico. In vivo the 3'-end being extended should anneal well without any mismatches or indels in the terminal 3-4 nucleotides since that may result in a "flappy end" that prevents the polymerase from docking to the duplex. The number of mismatches and indels allowed in the remaining part of the primer depends on the primer length, the melting temperature of the primer and the PCR conditions.

In silico it is possible to model the in vivo behaviour with a sticky 3'-end and a somewhat more loose annealing for the rest. I believe that is well implemented in the primer design software Primer3 (http://primer3.sourceforge.net/). However, pcr_seq ignores this. pcr_seq was developed to screen a template sequence for the primary PCR product, but also any side products arising from alternative binding of the primers in different configurations (forward vs forward and reverse vs reverse). I think it would be a good idea to start by allowing 2 mismatches, 1 insertion, and 1 deletion for a normal primer of length ~25. If this shows the desired product you can loosen the binding criteria by increasing the mismatches and then the indels until you see unwanted alternative products. Those you then adjust your PCR conditions to avoid.

I am CC'ing the Biopieces Google Group, so others may benefit/contribute.

Cheers,

Martin 

On Wed, Nov 20, 2013 at 12:54 PM, Salavirta Heikki <Heikki.S...@vtt.fi> wrote:

Hello,

You’re listed as the author of this piece of code. I was just wondering, do you have any suggestions for the number of allowed mismatches, deletions, and insertions so that the pcr_seq output would somewhat reflect real pcr on the given input. I’ve noticed that [3,2,1] definitely gives too many results for most primers and targets I’ve tested.

Thanks,

-heikki

-- 

Heikki Salavirta

VTT Technical Research Centre of Finland 
Geobiotechnology 
heikki.s...@vtt.fi 
+358406746364 
Tietotie 2

02044 VTT

Finland
www.vtt.fi