single primer PCR - (Jul/28/2011 )

I have a virus genome like this

5'-xxxxxxxxxxaccaacccattgaggttttgatgaccaaactgacttttactgccagggtactatggtccctagatt-3'

we don't know the sequence marked with x so we want to find out. Also, before 5' there are not bases because the genome of the virus starts from there.

I want to design only 1 reverse primer for the 3' part to amplify the fragment and then send for sequencing. Is this possible? Do I have to increase the number of PCR cycles?

Can't really see this 1 primer only idea working. I would be worried that in your case the unknown section is too close to the primer and has the poteintial to be a low quality or messy read.

What about using you use you reverse primer with a commercially available random hexamer (or something like this), clone the PCR product and then sequence a few clones? This enables a forward and reverse sequencing that begins far enough away from the unknown area to get a quality read.

The other method is RACE. The Race kits allow us to sequence the 5' and 3' ends of a fragment. But they are expensive, so I wanted to go ahead with single primer PCR, I've heard people getting it to work.

Is there some sequence after the 3' end you wrote or is that the whole virus genome? Because if so then design the reverse primer with a longer distance to unknown part, like 100 or 150 bases. If you have enough of the purified virus and the genome is not so big, you can directly sequence it.
If it's all there is, then you can't do direct sequenceing for reasons mentioned by Micro.

Other ideas:
You could design the reverse primer on the 3' end doing asymetric PCR. But since that will only cause linear amplification you would need like 100 cycles, and I'm still not sure if that would be enough for cloning. But you have to think about conditions, for example very short denaturation step to save the polymerase for such a long time and so on. Then clone it and sequence from vector.

Or just cut the genome to peaces and clone, then select those that are positive for your reverse primer+forward vector primer. But that requires lot's of selecting.

Or you can do RACE without kit, have a really unique-sequence oligo and ligase that will work on 5' ends (I suppose its a DNA virus, and not mRNA so it doesn't have a GTP cap and you can directly ligate it). Then you can use reverse primer and unique-sequence binding primer to have a PCR product, purify from gel and clone.

Hi Torf,

Yes there is some sequence after the 3' that I didn't write it down.

This is an RNA virus actually. I read about asymmetric PCR before. I could give it a try, but today my boss gave me the green light to buy Ambion's RACE kit. They say it works better than Invitrogen's RACE kit as it gives less smear and that the protocol is shorter.

However I am still very keen to do this PCR method later on. As I said, I have heard people getting it done.

Single primer PCRs should usually work, but at low speed, i.e. there's a linear not an exponential amplification. But I've no idea about results and how you can use it for sequencing....
RACE is a good idea insofar...