Re: [phenixbb] elbow

Hi Nat, So I added SMILES in Chemical string, but the conformation came out to be different than the input pdb file, even "none" was chosen for geometry optimization. Mengbin On Thu, Jun 13, 2013 at 3:20 PM, Nathaniel Echols wrote:

On Thu, Jun 13, 2013 at 11:53 AM, Mengbin Chen wrote:

...

I tried the command line in Terminal, but Elbow still regularized the conformation and bonds of the ligand and the result turned out to be the same with that I ran thorough GUI. The thing is that the input pdb file has a conformation the ligand should adopt so that I do not really want energy minimization or geometry optimization to be done on it. Although I set "none" for "geometry optimization" in GUI, it unfortunately changed the ligand conformation.

This has nothing to do with use of the GUI versus command line - the problem is that using a PDB file as the only source of chemical information is unlikely to work, especially if no hydrogens are present. At the very least you need explicit hydrogens, and ideally you need an input file that actually defines the topology, like SMILES or MOL2 or CIF. If it's a ligand that is already in the PDB you can just give it the three-letter code and it will extract this information automatically from the database included with Phenix.

I guess I need to add a warning to the GUI if users try to use a PDB file alone, because this problem comes up again and again.

-Nat

_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

-- Mengbin Chen Department of Chemistry University of Pennsylvania