Will rerun produce a different trajectory?

User discussions
,
1

GROMACS version: 2021.3
GROMACS modification: No

I created index to join groups of interest for gmx energy analysis. This requires a rerun from a trr file. I noticed that this approach produces a new trajectory file so my guess is they have different hash than the original trajectory. This must be because of the incorporation of index to the trajectory, but how about the positions and velocities of all atoms and molecules, are they entirely different? I am fine if its just within precision of error though.

2

You already have the original coordinates and velocities, so you can ignore the new trajectory file. The contents of the new file should only differ by rounding errors from the original one. I think that newer versions of GROMACS do not write velocities, but I’m not sure about that.