[gmx-users] trjconv -pbc

[Anton Feenstra]

Andre Farias de Moura wrote:
[...]
> gyration would not be correct, they would change every
> time a lipid molecule jumps across the boundary. is that
> right?
> 
> I think these programs should find out which molecules are
> actually in the aggregate before any structural calculation
> is performed... I just don't know how to do it yet.

Yes, you're absolutely right. The same problem can occur when
looking at a multimeric protein, or a protein with non-covalent
co-factor and/or ligand bound. This is one of the reason's I
implemented '-pbc nojump', on the basis that at the start of
your simulation you would generally have your 'aggregate'
located in the center of the box, i.e. 'whole' w/r to the pbc.
Then, eliminating any pbc jumps of all molecules in the
aggregate should keep it together. If that doesn't work in your
case, it might be due to extremely large frame intervals (in
the order of ns's I'd say for lipids), or else something is
broken.

If anyone would volunteer to 'wield the blunt axe' and re-organize
trjconv? I think that it has become terrybly convoluted due to the
additioon of the numerous options during the years... ;-)


-- 
Groetjes,

Anton
  _____________ _______________________________________________________
|             |                                                       |
|  _   _  ___,| K. Anton Feenstra                                     |
| / \ / \'| | | Dept. of Pharmacochem. - Vrije Universiteit Amsterdam |
|(   |   )| | | De Boelelaan 1083 - 1081 HV Amsterdam - Netherlands   |
| \_/ \_/ | | | Tel: +31 20 44 47608 - Fax: +31 20 44 47610           |
|             | Feenstra at chem.vu.nl - www.chem.vu.nl/~feenstra/       |
|             | "If You See Me Getting High, Knock Me Down"           |
|             | (Red Hot Chili Peppers)                               |
|_____________|_______________________________________________________|