GROMACS version: 2024.3-conda_forge
GROMACS modification: No
Hi everyone,
I am currently working on a project where I am trying to measure differences in my trajectory over time. I mainly use “rdf” and “hbond-legacy” (calling them using “subprocess” in Python).
I have a trajectory of 20,000 frames with 0.1 ps between each frame stored in an XTC file. Is there a quick way to analyze the whole trajectory all at once? Right now I iterate in the following manner:
-b 0.05 -e 0.15
-b 0.15 -e 0.25
-b 0.25 -e 0.35
etc.
This is incredibly slow, and takes approximately 5-6 hours. I was thinking that I could either write to scratch or split up my XTC file. Would any of this help or is there anything obvious I am missing? As soon as a frame has had its RDF and H-bond data generated and analyzed (by my Python protocol) it is safe to delete them.
Then I have another problem, not really GROMACS related, but when I call gmx frequently in this manner, at some point my Python SLURM job can not find gmx at the correct path if I run multiple SLURM jobs in parallel. I get a bash error stating that the gmx path does not exist, but it does. I guess there are too many calls, causing conflicts? If someone has a nice workaround for this too that would be appreciated.
Thanks in advance for any helpful pointers.
Best,
Alfred Nordman