Le 25 mars 2010 à 16:43, Tim Gruene a écrit :
> Hello Rebecca,
>
> this is merely a guess, but I think, it should work.
>
> As you calculate the electrostatic surface potential in pymol, it uses apbs.
> apbs, if I remember correctly creates an intermediate PDB-file where the
> B-factor column is replaced with the partial charge of that atom.
>
> So if you can get hold of that file you can use a text editor to cut out the
> helices of interest into separate files and use shell commands to sum up the
> B-factor columns.
>
> The following (all in one line, in case any mail program splits the line) adds
> up the B-values in the PDB-file:
>
> grep "^ATOM" helix.pdb|cut -c61-66|awk '{pcharge += $1}END{print pcharge}'
>
> Having said that I would be curious to hear what the community has to say
> about
> how trustful and meaningful such calculations actually are, because I have
> always wondered about this since I looked at my first GRASP surface.
>
> TimIt depends on what you want to do with the information. If you want something qualitative, like to decide whether a helix is more or less negative than another, it may be enough. But that's about all you can get from this approach. The coordinate file that you mention (in pqr format in fact) it's not unique and is not produced by APBS, but by other tools, such as pdb2pqr. That file will include charges and atom radii as derived from a particular force field. For surface representations, the PARSE force-field is a simple and perhaps good representation. In this force field the charges and atom radii are not the same than in other force fields, like CHARMM or AMBER. From this pqr file, APBS solves the Poisson-Boltzmann equation under certain conditions. Those include, for example, the presence of counter-ions in the implicit solvent. So, the results depend on the way you pose the problem, something that standard plugin users perhaps/probably don't do. In any case, I must confess that I don't understand what a "net charge of an electrostatic surface" means. In my mind, a net charge is a property of a molecule not of a surface. Actually, I would say that "electrostatic surface" is a shorcut for something like a "surface-mapped electrostatic potential". The electrostatic potential is not restricted to the molecular surface and its representation as an isosurface is often more informative than its mapping to a particular surface. These calculations can be very useful, for example to study binding properties, provided they are carried out carefully. Best, -- Miguel Architecture et Fonction des Macromolécules Biologiques (UMR6098) CNRS, Universités d'Aix-Marseille I & II Case 932, 163 Avenue de Luminy, 13288 Marseille cedex 9, France Tel: +33(0) 491 82 55 93 Fax: +33(0) 491 26 67 20 e-mail: miguel.ortiz-lombar...@afmb.univ-mrs.fr Web: http://www.pangea.org/mol/spip.php?rubrique2
smime.p7s
Description: S/MIME cryptographic signature
