Thank you Nigel, However, your cif file did not help with the problem (carboxylate is not planar) when using the PHENIX 1.4-3 version that we have now. You mentioned 2 newer versions of phenix, 1.4-4 and 1.4-6. Do I need to get them both? By the way, phenix.ready_set (in our current version) screwed up all hydrogens in my other ligand (trimer of ASG-GCU). I think that the ready.set needs to know that the monomers are linked. With this particular ligand that is a hexamer, ASG is in the library, but GCU is not. Should I give a cif_link file for the ready.set, too? After I get a newer version of phenix, I 'll try it again. Maia Nigel W Moriarty wrote:
Maia
Using PHENIX 1.4-4 and 1.4-6, eLBOW generates a CIF with 4 planar groups including the carboxyl group missing from the earlier PHENIX version I believe you used. I have attached it. There are a few other things I should mention.
I used a tool called ReadySet!
phenix.ready_set model.pdb
which will generate (using eLBOW) CIF files for all the ligands in the model using the data from the Chemical Components database from the PDB. I would recommend that you use ReadySet! for preparing a PDB file for refinement. It also adds hydrogens to your model which is usually a good thing.
Your example does have some nuances. You mention that it is covalently bound to the CYS. I have attached the two additional files needed to inform phenix.refine of the covalent bond. I used eLBOW directly for this but will ensure that ReadySet! will do the same function in the next release.
An additional function that I have been working on recently is the removal of the hydrogen in a carboxyl group when the ligand is in the model. I shall have that in the next PHENIX release.
Finally, I did a refinement using dummy reflection data and the carboxyl group remained planar.
If you have any problems with any of the files, feel free to contact me directly.
Nigel
------------------------------------------------------------------------
_______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb