[gmx-users] xdrf Fortran program and g_traj

[Mark Abraham]

Shaghayegh Vafaei wrote:
> Hi all,
> 
> I was wondering if somebody has worked or used the xdrf Fortran program in order
> to read the coordinates from a .xtc file. There are 6 errors after compiling
> the program . The subroutines are not included there and I couldn't find them
> in the links -> http://hpcv100.rc.rug.nl/xdrf.html ,
>          http://hpcv100.rc.rug.nl/xdrfman.html .
> 
> I am simulating a solution in a confinement between two parallel walls and I
> want to check how many molecules are adsorbed to the walls. Therefore, I am
> trying to write a program to keep track of the molecules coordinates.
> 
> Has anybody done something like this before? Did anybody use g_traj to read the
> coordinates? I can't understand what has been plotted, although I tried to plot
> z-component but the result of z-component is the same as that of x-component.

If you read g_traj -h, then you'll have a better idea about why it isn't 
suitable.

If you're looking to count the number of molecules adsorbed on another 
group, g_mindist will probably work for you. Define a group for the wall 
and each relevant molecule as a group.

If you're really desperate to write your own code, I suggest using 
GROMACS trajectory-reading machinery. Look inside src/contrib/README. 
Choose the analysis program that seems closest to what you want to do, 
copy relevant bits to src/contrib and modify that, rather than engineer 
from the ground up. Read the manual chapter 8 to find out what all the 
analysis programs do, then look for more detail with "g_xxxx -h" or "man 
g_xxxx" or further in the manual.

MArk