Re: [ccp4bb]: Comments on ligands

[Ethan Merritt <merritt@u.washington.edu>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

On Friday 15 November 2002 08:54, Gerard DVD Kleywegt wrote:
> a model deposited without data (or not at all) one has to take on
> faith (i.e., more religion than science). for a model deposited with
> data, you can go and check the electron density yourself (e.g., using
> the uppsala electron density server, EDS, a pre-release version of
> which is available at http://fsrv1.bmc.uu.se/eds).

Certainly you can. But the issue being raised in this particular case
was not the fit to density, but the internal stereochemistry of the ligand.
I think it is a valid point and maybe shows the need, or at least the
utility, of a generalized validation tool that looks at all stereochemistry
in a deposited model, not just that of the protein.

There is plenty of room for honest mistakes in fitting unusual ligands
to binding sites in a protein.  I try to make sure that I run all such fits by
one of the local chemists before investing time in biological interpretation
or deposition, and I freely admit I've been told more than once that the
initially fit conformation did not look chemically plausible.  Then ensues
an attempt to refit the density with another conformation, and at the same
time a hunt through the Cambridge data base for similar small molecule
structures plus maybe even some ab initio energy calculations.

-- 
Ethan A Merritt       merritt@u.washington.edu
Biomolecular Structure Center Box 357742
University of Washington, Seattle, WA 98195
phone: (206)543-1421
FAX:   (206)685-7002