[phenixbb] DNA double helix with many molprobity bad clashes

[Nathaniel Echols]

On Fri, Nov 30, 2012 at 3:14 AM, Morten Groftehauge
<mortengroftehauge.work at gmail.com> wrote:
> Check that your OP1 and OP2 aren't swapped. Phenix doesn't check for this
> and is unable to resolve it automatically but your geometry gets distorted
> if they are swapped.

Following up on this (we had considerable discussion off-list),
apparently the problem was an issue of nomenclature - some of the
hydrogen atoms were not being recognized by Probe as covalently bonded
to the DNA backbone, so it thought there were clashes there.
(phenix.refine is more flexible about atom names because Ralf added a
bunch of special cases to the PDB interpretation - but Probe has an
entirely monolithic codebase.)  What we haven't been able to do yet is
reproduce the problem using only tools in Phenix, so if anyone else
has come across the same thing, we'd be very interested in seeing the
model with a description of how it was obtained.

-Nat