Hello I am experiencing the same problem as the person who emailed this initially (pasted below; the entire thread that I found in the archive is pasted after my email, for reference).
*Hello, * *> I modelled a small molecule with the dundee prodrg server * *> and can read the pdb without any problems into Coot. But * *> when I try to import the cif.dictionary I only get the * *> message: ooops, cif dictionary contains no restraints. (to * *> create this file I copied and pasted the refmac5 file from * *> prodrg into a new file). * *> Can anyone please explain to me what this error message * *> means? * Reading the responses, it appears that there are some empty loops in the cif file generated by proDRG that this user had to remove and all was well. However, I looked at my cif file and none of the loops are empty. When I compare this file to one that coot accepts, the only difference is that the coot-accepted version has xyz values for the ligand: loop_ _chem_comp_atom.comp_id _chem_comp_atom.atom_id _chem_comp_atom.type_symbol _chem_comp_atom.type_energy _chem_comp_atom.partial_charge _chem_comp_atom.x _chem_comp_atom.y _chem_comp_atom.z The last three lines are absent in proDRG cif file. I removed these lines as well as the coordinates from the coot-accepted version but coot reads the file anyway. Clearly, this isn't the problem. What could it be? I am unable to use sketcher/libcheck or a refmac5 generated cif file for the ligand. Sketcher/libcheck won't work (but that is a different problem, which I have to fix separately, as sketcher doesn't recognize even a plain old cyclohexanol), and refmac5 adds and deletes double bonds randomly, and messes the geometry up completely. Any help with the proDRG cif file will be appreciated!! Thanks, Sangeetha. From: Paul Emsley <> Date: Thu Dec 01 2005 - 04:29:05 GMT On Wed, 2005-11-30 at 19:14 -0500, Joel Bard wrote: *> Usually this means that there is some sort of little error in the * *> format of the cif file and it gives up reading before it gets to the * *> restraints. * Exactly. Mareike send me her cif file and there was indeed a syntax error in it. The thing to look for is "dirty mmCIF file? xxx.cif" in the console. We've had this problem before in fact. IIRC we contacted the authors of PRODRG about it - they seem not to have fixed the problem yet. To fix PRODRG cif files then: At the end of PRODRG cif files we can get empty geometry description loops. In this case we have: loop_ _chem_comp_chir.comp_id _chem_comp_chir.id _chem_comp_chir.atom_id_centre _chem_comp_chir.atom_id_1 _chem_comp_chir.atom_id_2 _chem_comp_chir.atom_id_3 _chem_comp_chir.volume_sign loop_ _chem_comp_plane_atom.comp_id _chem_comp_plane_atom.plane_id _chem_comp_plane_atom.atom_id _chem_comp_plane_atom.dist_esd # # --------------- # This causes the mmdb's mmCIF parser to choke. Simply remove these lines and the file can be parsed. Paul. *> You can try manually comparing it with a cif file that coot likes. An * *> easier thing to do might be to read your monomer (as a pdb) into the * *> ccp4 monomer sketcher and have it generate a new cif file. Coot * *> usually likes the cif files from the monomer sketcher. * *> * *> Good Luck, * *> * *> Joel * *> * *> >>> "Mareike Kurz" <[email protected]> 11/30/2005 5:41:26 PM >>> * *> * *> * *> Hello, * *> I modelled a small molecule with the dundee prodrg server * *> and can read the pdb without any problems into Coot. But * *> when I try to import the cif.dictionary I only get the * *> message: ooops, cif dictionary contains no restraints. (to * *> create this file I copied and pasted the refmac5 file from * *> prodrg into a new file). * *> Can anyone please explain to me what this error message * *> means? * I suppose that such a question means that i) it should be in the FAQ and ii) I should propagate the dirty cif message to the gui. Hmmm...
