How to avoid PCR errors?

[Frederick Roth]

In article <oliver.4.2D5A3C2F at molgen.biologie.uni-marburg.de> oliver at molgen.biologie.uni-marburg.de writes:
>Dear netters,
>I want to amplify 2 kb genomic DNA and then to subclone the DNA (actually I 
>have to amplify it two times, because I perform a site directed mutagensis). 
>It is essential, that the sequence of the subcloned DNA is the SAME as the 
>original sequence. Until now I did not manage to get a clone without PCR 
>errors.
>Has anyone experience with such a problem and knows how to circumvent this? 
>I think there are some polymerases wich have a proof reading function. Has 
>anyone ever used such an enzyme and can tell me wich is the best to use for 
>this purpose?
>
>Any help is highly appreciated,
>
>Oliver
>==========================================================================
>     Novell-Server Molekulargenetik Biologie Uni-Marburg Deutschland
>==========================================================================


Check out Pfu....  Lundberg et al., Gene 108:1-6 (1991)

Pfu (from Pyrococcus furiosus-- 'rushing fireball') is a proofreading 
polymerase w/ 1 or 2x10^-6 error rate.