Qiagen: This is ridiculous

Oliver S. DePeyer saspeyer at reading.ac.uk
Thu Aug 29 09:27:14 EST 1996

I recently had a lot of problems with my gravity-fed Qiagen minis... I
think the problem is that the time intervals specificied in the protocol
are too stringent. Basically, every time I have tried using only one or
two columns at a time, and have timed all the steps more or less to the
second, it has worked fine, with 100s of ng yield. BUT, when I've tried
using >5, 6, 7 columns etc. at once, then usually I find that maybe a
couple work and the rest don't. The only difference I can see is that
whereas for 1 column on its own, 5 minutes between steps really IS 5
minutes, with more columns it is 5 minutes for the 1st column, 5 mins 20
secs for the second, 5 mins 35 for the third... even if you are a fast
pipetter. And the further the column is down the order, the more erratic
the results are. The 1st one, where it's been closest to the protocol
timing, usually works best! It's very hard to temporally stagger the
different columns though, especially the centrifuge run at the start - 
any reaosnable person would put all the samples in at once. And
since an entire Qiagen gravity-fed run takes about an hour, you could
easily spend a whole day doing them all one after the other!  I agree that
this really isn't robust enough. But, on the other hand, many automated
sequencers such as ABI really aren't that happy (in my experience) with
templates purified by any other method... 

More information about the Methods mailing list