Is it possible that the imaginary object/atom (to which the spring is attached, and which moves at a constant velocity - shaded in grey in the image) passes through the PBC? If it does and appears on the other side of the box, does the reaction coordinate change?
If I am not mistaken, immediately after passing the PBC an error will occur because "pull distance is greater than half of the box size".
You may have to increase the box lenght or use absolute coordinates to avoid this.
Dear Dheeraj Prakaash
It should depend on how you define the reaction coordinate. If you use, as classically done, the distance among COMs, then after going 0.49 times the box length you will get an error. This should be related to the fact that if you have PBC on, GROMACS uses the nearest image to calculate distances. Once your position is more than half the box length, the COM distance becomes ill defined.
You can avoid this by specifying direction-periodic as pull type in the mdp options. Take a look here
https://manual.gromacs.org/documentation/2018/user-guide/mdp-options.html#com-pulling
However you must set to zero the compressibility along the axis that is your reaction coordinate. Say you are trying to pull a lipid from a membrane with the transmembrane axis z as reaction coordinate with the distance from the bilayer COM as reference, and you use a semi-isotropic pressure coupling, the compressibility along z must be zero. In this way you should avoid the problem of a COM ill-defined and be able to pull along that direction for how much you need.
- Badges
- Science topic
- Similar topics
- Chemistry
- Theoretical Chemistry
- Computational Chemistry
- Molecular Dynamics