Re: [phenixbb] rotamer outlier reduction by coot

Hi Dialing, I agree with Nat, you probably have incompatible hydrogen names, which is what causes the hydrogens to "explode". I'm not sure what you mean by oxygens packing together. Do you mean that there are oxygen atoms that are too close together? It's difficult to know for sure what the problems are without seeing the files first hand. If you send me your starting PDB file, and the file that you saved out of Coot, I will have a look. Please send them directly to my email address, NOT to the list. Jeff On Mon, Dec 5, 2011 at 10:54 AM, Nathaniel Echols wrote:

On Mon, Dec 5, 2011 at 2:48 AM, Dialing Pretty wrote:

...

After select a rotamer and acception, I find some hydrogens in the outlier rotamer will be isolated from the rotamer, and the rotamer will have new hydrogens at its new position. Is any influence by further refine by phenix?

This sounds like the classic "exploding hydrogens" problem.  You should update to the latest version of Coot (possibly one of the nightly builds), which has the new hydrogen definitions which are more compatible with Phenix.

...

After I save the new PDB file by coot after I selected the new rotamers to replace the outlier rotamers, I did a Phenix refine. There is an error message at the start says "Number of atoms with unknown nonbonded energy typensymbol: 2228". What does it mean and how to solve the problem?

We can't answer this accurately without knowing more about what atoms it complains about.  Normally this indicates that restraints are missing, but it's possible this is related to the problem with hydrogen atoms.

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