Many thanks for the reply.
I forgot ptraj can just deal with PSF files or PDB files instead of a
prmtop!
That's really great, BUT:
I've tried this with a 'simple' example and I can't get ptraj to output the
correct distance between to atoms.
The atomic coordinates for the distance I tested are:
ATOM 1328 CA GLU 82 5.338 -15.135 0.848 1.00 0.00
ATOM 3542 CA THR 218 1.905 -18.135 -5.190 1.00 0.00
The correct distance is thus (about): 7.566
I ran ptraj as follows:
ptraj ionized.psf dist_test.in &> dist_test.out
with dist_test.in containing:
trajin ionized.pdb
distance test out dist_test.dat :94.CA :230.CA noimage
dist_test.dat contains the following incorrect distance: 19.847652
As the residue numbering in ionized.psf starts at 13 instead of 1, I also
tried the equivalent (with the residue numbers in the masks specifying the
numbering starting from 1):
trajin ionized_start1.pdb
distance test out dist_test.dat :82.CA :218.CA noimage
now dist_test.dat contains the following incorrect distance: 14.942543
I tried just writing a PDB with ptraj as well, and that works, i.e. I get
the same residue numbers and coordinates as specified in the trajin .pdb
file.
Further relevant information: I'm using ptraj from AmberTools1.5 with
bugfixes up to bugfix.13
Any insights as to why ptraj is giving me an incorrect distance?
Further, when I tried using the PDB file (ionized.pdb) as prmtop, I got a
segfault and errors in the ptraj output that indicate it is expecting a
prmtop instead of a .pdb:
Many thanks in advance for any help/insights!
Marc
On 15 September 2011 13:21, Thomas Cheatham <tec3.utah.edu> wrote:
>
> > As I'd like to use ptraj to analyze someone else's trajectories (run
> using
> > X-PLOR .psf files in NAMD), this feature would be really useful for me.
>
> ...for much analysis in ptraj you can get away with using a PDB file
> instead of a prmtop (and/or the .psf). Ptraj "should" handle all PSF
> files and if not I need to fix it.
>
> However, for non-ptraj (i.e. MM-PBSA) then getting CHAMBER working is
> appropriate (and shouldn't be too hard).
>
> --tec3
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Sep 15 2011 - 06:30:03 PDT
