One way to check if your trajectory is truncated or otherwise has issues is
to calculate the expected number of lines for each frame in the trajectory
file:
#Lines = (#atoms * 3) / 10
Round up to the nearest integer if there is *any* remainder to the division,
and add 1 more if you have box coordinates (which appears to be the case
here). So for example, a system with 332 atoms and box coordinates would
have 101 lines each frame. Then just multiply #Lines by your expected number
of frames, and see if that number matches the output of a 'wc -l' command.
In my above example if I had 100 frames in the trajectory I would expect the
file to have 10100 lines - any less or more and I know I've got a problem
somewhere.
If your trajectory is corrupted ptraj cannot currently jump to the next
non-corrupted frame on its own. You will need to manually find the first
'good' frame after the corruption and separate the trajectory file.
One way to avoid this in the future is to use the NetCDF trajectory format
(ioutfm=1). Even if some frames become corrupted in a NetCDF file there is a
good chance the other frames will be ok.
Hope this helps.
-Dan
>
>
>
> 2010/4/12 Jason Swails <jason.swails.gmail.com>
>
> > How is it corrupted? Did you use iwrap=1 in your simulation input
> > file? If not, I can imagine that starting at frame 7.659, one of the
> > water molecules has diffused too far for the fixed-format output of
> > the mdcrd to handle (and thus prints ***** for the coordinate). If
> > this is the case, then every frame after 7.659 is probably corrupted.
> >
> > It would also be helpful to see your ptraj script when you tried to
> > read in, for instance, every 5th frame. You may consider trying to
> > read in the first 7.658 frames and printing those out and visualizing
> > them to see what may be going on (i.e. if it is the diffusion problem
> > i mentioned).
> >
> > Use the following trajin command:
> >
> > trajin inputfile.mdcrd 1 7658 1
> > strip :WAT:Cl-
> > trajout outputfile.mdcrd
> >
> > In the future, I suggest you try using netcdf trajectory files instead
> > of ascii mdcrd files (see variable ioutfm in the amber manual to use
> > this format in both sander and pmemd). Ptraj can read these files as
> > well, and they are faster to write, read, and process, they hold
> > higher precision numbers, and they are a fraction of the size. Also,
> > if you do not have AmberTools1.3 installed, this version of ptraj may
> > be useful in this problem (it only reads frames you intend to process,
> > and so is much faster and will skip bad frames if you don't include
> > them in analysis).
> >
> > Good luck!
> > Jason
> >
> > On Mon, Apr 12, 2010 at 5:08 AM, Silvia Carlotto
> > <silvia.carlotto.unipd.it> wrote:
> > > Dear amber user,
> > >
> > > I have a big trajectory .mdcrd file (29G and 500.000 frames) and I want
> > to
> > > strip water molecules.
> > >
> > > The preview trajectory (29G) do not give me any problem but the second
> > one
> > >
> > > give me this error in ptraj
> > >
> > >
> > > Warning in readAmber trajectory():Set #7659 is corrupted (3.)...
> > >
> > >
> > > The .out file does not give me some kind of errors and I used the
> restart
> > > file of
> > >
> > > this simulation in other simutations without any errors.
> > >
> > >
> > > My ptraj scrip is simply
> > >
> > > trajin inputfile.mdcrd
> > > strip :WAT
> > > strip :Cl-
> > > trajout outputfile.mdcrd
> > >
> > > I try to:
> > >
> > > 1) save every 5 frames to jump the damege frame
> > > 2) cut the trajectory and save the part after the corrupted frames
> > >
> > >
> > > But every time the ptraj tool stop at #7659 frame.
> > >
> > > The time mahcine calculation is very long so I want to know if I can
> save
> > in
> > > some way my simulation.
> > >
> > > Is it possible to jump in same way the corrupted frame or to delate
> from
> > the
> > > mdcrd file?
> > >
> > > The file is too big to open it with a text editor.
> > >
> > > Plese help me!!!
> > >
> > > I have not other idea!!!!
> > >
> > >
> > > silvia
> > > _______________________________________________
> > > AMBER mailing list
> > > AMBER.ambermd.org
> > > http://lists.ambermd.org/mailman/listinfo/amber
> > >
> >
> >
> >
> > --
> > ---------------------------------------
> > Jason M. Swails
> > Quantum Theory Project,
> > University of Florida
> > Ph.D. Graduate Student
> > 352-392-4032
> >
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
>
>
>
> --
> Silvia Carlotto, Ph.D
>
> Università degli Studi di Padova
> Dipartimento di Scienze Chimiche
> via Marzolo 1, 35131 Padova
>
> Telefono: +390498275124
> E-mail: silvia.carlotto.unipd.it
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
-------------------------
Daniel R. Roe
Postdoctoral Associate
SAS - Chemistry & Chemical Biology
610 Taylor Road
Piscataway, NJ 08854
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Apr 12 2010 - 06:00:07 PDT
