Hi Oliver.
> > PATCH RESIDUE_TYPE = 'HS2', RESIDUE_IDS = 'His number' >
I didn't reply to your previous post, because I wasn't sure what the quickest suggetion would be. However, it looks like your getting there yourself.
You can learn a lot about the cryptic errors that modeller generates by examining the nomenclature that it uses for atom and residue types. You can find out about these from files in the${MODELLER6V2}/modlib directory. The key one is restyp.lib, and the particular parameter and topology files you are using (eg top.lib and par.lib contain the topology and atom interaction parameters, respectively, corresponding to the modeller types translated from the table in restyp.lib).
You might also want to turn on'extensive' logging( "SET OUTPUT_CONTROL = 1 1 1 1 1" ) to find out when these warnings are being generated. The manual can be quite illuminating once you've found out which particular function is complaining!
For the ACE/CT3 patches, it looks a little like you haven't turned off the default patching: http://salilab.org/modeller/manual/node29.html#3159 As for the water model, I'm not sure exactly, since it looks like the correct class is associated with the OH2 atom, but have you got WATER_IO set correctly ? Bear in mind that HETATMs are distinct from waters, so different functions are applied in order to build them into the model.
with any luck, some of the above are actually hints ! j. _______________________________________________________________________ Dr JB Procter:Biomolecular Modelling at ZBH - Center for Bioinformatics Hamburg http://www.zbh.uni-hamburg.de/mitarbeiter/procter