[gmx-users] Re: Re: rerun of trajectory how to deal with the huge .trr file?

Dallas Warren dallas.warren at vcp.monash.edu.au
Tue Mar 9 00:22:01 CET 2004


>1/ when runing mdrun,how can i keep water molecule out of the output 
>file(such as.trr)   in order to minmize the size of the outpur file

You can't during the run with .trr files (I don't think you can, haven't 
seen anything that indicates you can).  Afterwards, you can use trjconv to 
remove the water, but not while the simulation is running etc.

>2/ according to what you said in gromacs mail lists: "if you use the .xtc 
>file format then you don't have to save to ..trr very often at all, only 
>really the final step if you want to extend the simulation.  This can be 
>wasteful though, as if your simulation crashes somewhere then you have to 
>start from the very start again and there can go a significant amount of 
>CPU time.  Personally I have mine save every about 6 hours real time and 
>only keep the last .trr file so if the need arises I can continue the 
>simulation.  Everything else is saved and archived as .xtc"
>i do not understand what you said above.if  you only use the .xtc file as 
>output file,how did you  save the position and velocities to .trr  file? 
>if you use both .trr and .xtc file as out files,how did you divide the 
>.trr file every  6 hours when mdrun was running? terminated the mdrun 
>,delete the .trr file every  6 hours or other method? This point confuse 
>me most,please give me a detailed interpretation.  thank you!!!

I use both *.trr and *.xtc.  The *.trr file is used to continue or extend a 
simulation, *.xtc is used for the visualisation and analysis.  The *.trr is 
saved to about every 6 hours (500ps), *.xtc at around every 5 ps or 
something like that.  Everyone has different priorities etc in the 
information they are gathering, so do different things.

How often you write out to both of those files is defined in the Output 
Control section of the md.mdp file.  In particular for the *.trr file 
nstxout for coordinates, nstvout for velocities and nstfout for forces (if 
interested in those).  For *.xtc it is nstxtcout for coordinates and 
xtc_grps for the groups that you want in the output.  In that last bit you 
can use that to output without water if you want, but I always make sure I 
have a copy with all waters present.  Again, depends what is the priority 
with the information you are gathering etc.

>3/ How to use trjconv remove all the atoms that are not of importance 
>to  see fromthetrajectory, such as  water/SOL.i want to know which option 
>trjconv command used can function.please give me a command line.

trjconv -f input.xtc -o output_nowater.xtc  -n index.ndx

The index.ndx file just has to contain the appropriate group of atoms that 
you want to be in the output trajectory.  Generate the index file using 
make_ndx.

Catch ya,

Dr. Dallas Warren
Research Fellow
Department of Pharmaceutical Biology and Pharmacology
Victorian College of Pharmacy, Monash University
381 Royal Parade, Parkville VIC 3010
dallas.warren at vcp.monash.edu.au
+61 3 9903 9083
--------------------------------------------------------------------------
When the only tool you own is a hammer, every problem begins to resemble a nail.




More information about the gromacs.org_gmx-users mailing list