Hi all,
we're trying to solve a molecular replacement with phaser with data at 3.8 angstrom.
We only have half of the model (of a different species) to search for, and have to look for four models in the asymmetric unit (not an straightforward case, indeed).
The thing is, we start with not-so-nice Z scores for the first molecule, then when the program tries to fit the second molecule, we do obtain nice Z scores (even higher than 20) but with a huge number of packing clashes (not less than nine - and indeed the models superpose if we take a look at those solutions).
So my question is: are these Z scores indicating we're somehow near to a good solution or do we have to consider as good only the (high) Z scores of those solutions which show very few clashes?
Thanks a lot in advance,
Sebastiano
Sebastiano Pasqualato, PhD
IFOM
Istituto FIRC di Oncologia Molecolare
via Adamello, 16
20139 Milano
Italy
tel +39 02 574 303 325
fax +39 02 574 303 310
