PCR from genomic DNA

[krishnan at BORCIM.WUSTL.EDU]

Hi everybody!
a couple of weeks ago I requested the netland on ideas to PCR-amplify a 3 
kb DNA from human genomic DNA/total yeast DNA containing the YAC clone.  
Among the suggestions we  got were:
(i) to vary Mg conc, (ii) to use as little as 50-100ng of total DNA, (iii) 
lower/higher annealing temperatures, (iv) very short denaturation time 94 C 
for 5 sec, (v) inclusion of a final cycle at 72 C for 10 min.  We tried all 
these on the abovementioned DNAs and on the cosmid clone also - in vain.
	Finally, we obtained pBluescript clones of the cosmid and tested 
these in PCR assays using the same set of primers and the following 
conditions: ~1 ng DNA, 20pmol each primer, 200um dNTPs, 2.5mM Mg and 1U
Amplitaq.  The cycling conditions were: 94 C 1 min, 55 C 1 min, 72 C 4 min 
for 25 cycles with a last cycle 10 min extension step.  Voila! I got the 
expected 3 kb product.  We also tested this clone (contained a 4 kb insert) 
with modified universal and reverse primers in PCR under the same 
conditions and obtained a 4 kb PCR product.  The products ought to be 
specific because when one of the primers was left out, no product was 
obtained.  Note that the Bluescript clone was not a mixed population, but a 
specific 4 kb clone containing the 3 kb piece we were interested in.
	At least for the above set of primers and for this portion of the 
genome, it seems the length of the product to be amplified may depend on 
the complexity of the original template.  It surprises me that the cosmid 
template failed to yield a product.  Would the netters like to comment?
	Thanx v. much again to all those who shared their time and 
suggestions.
	Raja