Protein_ligand.ipynb

[zjun028]

I am currently utilizing the Protein_ligand.ipynb. However, I have encountered a notification from Colab indicating that the GPU is not being utilized during the runtime. Additionally, I am observing a simulation efficiency rate of approximately 95ns/day.

Could you please confirm if this efficiency rate is within the expected range for the given setup? Furthermore, I would appreciate any advice on how to ensure that the GPU is being actively employed during the simulation to potentially enhance the computational speed.

Thank you for your assistance and I look forward to your valuable input.

[ysa123-2]

I am also facing the same problem. A few days ago I subscribed to google colabs which allows me to running 24 hours. but with a small GPU capacity. As a result, I tuned my molecular dynamic with 50 ns for 10 hours, always encountering errors.

I repeated it again, still in the same condition, namely the error in the middle of the road.

Can you give an explanation/confirm of how much GPU capacity is needed and what is the maximum running time (I mean, the length of running (regarding nanoseconds) and waiting time (regarding total length, for example 10, 20 or 24 hours)?

Thank you for your assistance and I look forward to your valuable input.

[zjun028]

After completion, the analyze result: Interaction Energy Average = 230126125.39 ± 938448224.75 kcal/mol

Is there something wrong with the "Interaction Energy Average", it is so big value.

[noumanistic]

After completion, the analyze result:

Interaction Energy Average = 230126125.39 ± 938448224.75 kcal/mol

Is there something wrong with the "Interaction Energy Average", it is so big value.

Sir try saving trajectories in pdb format instead of dcd.

[polaris722]

hi pablo During the execution of the install dependencies stage on line 21, the following error is encountered: ModuleNotFoundError: No module named 'openff'. Where could the problem have originated from, and how can it be resolved?