[phenixbb] Planarity restraint for covalent linkage

[Nathaniel Echols]

On Fri, Jul 23, 2010 at 2:16 PM, Elwood A. Mullins <mullins at purdue.edu>wrote:

> I'm having issues defining planarity restraints in a .edits file. When all
> four of the atoms in the plane are selected in a single planarity entry,
> refinement crashes. When I define a single atom in separate planarity
> restraint entries (as suggested by Nat), refinement doesn't crash, but the
> four atoms in the plane are treated as four one-atom planes (obviously
> useless). My previous correspondence is attached below.
>

The reason it was crashing is that the atom selection parameter was
misnamed, and was being omitted from the final parameters - but since you'd
set the sigma, it still tried to create a restraint with an empty selection.
 After taking another look at the corrected file I sent you, I see I didn't
read it closely enough.  The atom selection for each plane needs to contain
*all* atoms in the plane - unlike the bonds and angles, you do not define
individual atoms separately.  The correct syntax is actually this:

    planarity {
      action = *add delete change
      atom_selection = "(chain B and resid XX and altid B and (name CD or
name CG or name OE1)) or (chain C and resid YY and altid B and name S1P)"
      sigma = 0.02
    }

(I've deliberately obfuscated the residue numbers, but you get the idea.)

Sorry for the confusion - I was at a conference and attempting to multitask.

-Nat
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20100724/ad45493a/attachment-0002.htm>