VMD-L Mailing List

From: lloyd riggs (lloyd.riggs_at_gmx.ch)
Date: Sun Sep 30 2012 - 13:08:40 CDT

Dear All,

I signed up and wrote last week about me switching to use VMD/NAMD to look over gromacs trajectories.

I have now hit this point.

I can with 130K water molecules, cut them into 10K pieces and generate a separate .psf for each. I have no problem with the small molecules, solvent and protein portions.

The problems I have are concatenating these, as the waters have different chains, but the same residue number as this fortran like column only goes from 0 to 9999 , and then VMD complains of duplicate residues. Theres no way to parse in a letter, like 2A45 which would at least allow for 800K or so.

In any case, the other problem is with atoms, this column only allows up to 0 to 99999 and then complains about duplicate atoms.

Is there a way to either concatenation a psf file properly, or use something like a gromacs system where for solvent you read in 1 type definition (say TIP3) with the angles and bonds, and then it applies these to all coordinates, thus saving the space, and duplicate problem?

I also tried reading them in separate, for the same large pdb, but then I get an incorrect number of atoms, or your not using x-plor and it misses the first column error.

Sincerely,

Stephan Watkins
University of Bern
Department RIA
Sahli Häus 2 Rm 43
Freiburgstrasse 22
BERN 3011 CH