how to run "run_prediction.py" in a alphafold docker instance?
Opened this issue · 7 comments
I have alphafold 2.3.2 installed and running using: python3 alphafold/docker/run_docker.py ...
but I can't figure out how to run TCRdock's run_prediction.py "in the alphafold docker instance"
Apologies for my lack of familiarity with docker.
Thanks, Keen to give this a try.
I think I replied to this by email but let me know if that's not the case! If it works we can add a description here...
Hi there!
Yes, I actually built a Docker container specifically for TCRdock, you can find it in the docker folder here in the repo.
You have to cd into the docker folder where you can then do docker build -t tcrdock . to build the container.
You can then run the container using the docker run command, the name of the image will be tcrdock. In this documentation, you can find instructions as to how to mount folders into the container so that you can access your input/output files.
Once you are in the container you can run the TCRdock python files via python /opt/TCRdock/run_prediction.py for example.
Hope this helps, if there are further questions just ask :D
Oh, and one more thing: If you want to use GPUs for the predictions, you need to perform steps 1 and 4 as described in the alphafold repository. You then need to use the --gpus all flag in you docker run command.
Awesome! Thanks so much, Elias. Perfect timing for this request!
…
________________________________ From: Elias Ball @.> Sent: Friday, March 8, 2024 7:21 AM To: phbradley/TCRdock @.> Cc: Bradley PhD, Phil @.>; Comment @.> Subject: Re: [phbradley/TCRdock] how to run "run_prediction.py" in a alphafold docker instance? (Issue #14) CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you recognize the sender and know the content is safe. Hi there! Yes, I actually built a Docker container specifically for TCRdock, you can find it in the docker folder here in the repo. You have to cd into the docker folder where you can then do docker build -t tcrdock . to build the container. You can then run the container using the docker run commandhttps://urldefense.com/v3/__https://docs.docker.com/reference/cli/docker/container/run/__;!!GuAItXPztq0!lr4PuCc4xCmeEWjxnSYJukiPXFFj_FHUBK8RY9peCPYPJGaW16fUQxK9B_Gvq2AlTNU-xDGrDQ87W2zb3MFx9sM-$, the name of the image will be tcrdock. In this documentation, you can find instructions as to how to mount folders into the container so that you can access your input/output files. Once you are in the container you can run the TCRdock python files via python /opt/TCRdock/run_prediction.py for example. Hope this helps, if there are further questions just ask :D — Reply to this email directly, view it on GitHubhttps://urldefense.com/v3/__https://github.com/phbradley/TCRdock/issues/14*issuecomment-1985883033__;Iw!!GuAItXPztq0!lr4PuCc4xCmeEWjxnSYJukiPXFFj_FHUBK8RY9peCPYPJGaW16fUQxK9B_Gvq2AlTNU-xDGrDQ87W2zb3CuPUwEY$, or unsubscribehttps://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/ABBNCH4J2SZS5WPV7T7CUJDYXHJPLAVCNFSM6AAAAABEMDDUCOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSOBVHA4DGMBTGM__;!!GuAItXPztq0!lr4PuCc4xCmeEWjxnSYJukiPXFFj_FHUBK8RY9peCPYPJGaW16fUQxK9B_Gvq2AlTNU-xDGrDQ87W2zb3C_3Uogx$. You are receiving this because you commented.Message ID: @.***>
Of course, happy to help :)
Thanks so much! I got the docker container to work. Running it like this:
docker run --gpus all -it
--mount type=bind,source=/path/to/tcrdock,target=/path/to/tcrdock
--mount type=bind,source=/path/to/AFdatabase,target=/path/to/AFdatabase
tcrdock bash
Then I run the python programs as per the instructions. If I'm doing something dumb, let me know. I'm not that familiar with this stuff.
@patycake57 I am doing something similar; just note that depending on how the permissions are configured in your host filesystem directories, you may not be able to write output to the mounted directory from within the container itself. Let us know how you get on!