[phenixbb] Hirshfeld (rigid bond) test

amakal at chem.uw.edu.pl amakal at chem.uw.edu.pl
Thu Jan 31 02:42:05 PST 2013


Hi Nat,

thank you for responding.
I am looking forward to the next build.

> We certainly have the ability to do this easily, and I would like to
> add more analysis features to the GUI anyway, but it's not clear to 
> me
> how to present the results for a macromolecule.  I threw together a
> script to expose these functions, which I'll call
> "mmtbx.rigid_bond_test" (it will be in the next build - I can also
> send you the script but I had to change another file as well).  The
> output looks like this:
>
> =============================== Rigid-bond test 
> ===============================
>
>   pdb=" N   LEU A   4 " pdb=" CA  LEU A   4 "     24.957
>   pdb=" CA  LEU A   4 " pdb=" C   LEU A   4 "     61.442
>   pdb=" CA  LEU A   4 " pdb=" CB  LEU A   4 "     19.558
>   pdb=" C   LEU A   4 " pdb=" O   LEU A   4 "     75.078
>   pdb=" C   LEU A   4 " pdb=" N   THR A   5 "    112.578
>
> ...and so on, but that doesn't seem terribly useful to me by itself.

I guess it is not terribly useful in most cases, but it would be a 
valuable tool for
comparing the outcomes of different refinement strategies for the same 
macromolecule,
especially as rigid-bond-test, unlike ADP values, should be in 
principle independent
from effects of different anisotropic scaling factors.

Averaging rigid-bond-test values over domains, main chain/side chain of 
a certain residue,
or over certain bond-types (peptide, C-Cring a.s.o) would be a nice 
validation tool too.

> It would be nice if there was a way to visualize this in PyMOL but 
> I'd
> have to think a little more about how to do that.

I am not terribly familiar with PyMOL, but I think that coloring bonds 
by the actual
rigid-bond-test value would be just great - I guess I would have to 
think about it too.

With regards,

Anna Makal



More information about the phenixbb mailing list