Zuricho / ParallelFold

Modified version of Alphafold to divide CPU part (MSA and template searching) and GPU part. This can accelerate Alphafold when predicting multiple structures
https://parafold.sjtu.edu.cn
145 stars 45 forks source link

How to run GPU part? #10

Closed hrzolix closed 2 years ago

hrzolix commented 2 years ago

How do I run model inference on GPU part of the process after featurization step? Does the model inference step automatically find feature.pkl in some folder?

Zuricho commented 2 years ago

You're right, run_alphafold.py without feature flag will automatically find feature.pkl in output folder, if feature.pkl exists, it will skip featurization step.

hrzolix commented 2 years ago

So basically I ran this: run_alphafold.sh -d $ALPHAFOLDDB -o test_parafold -m model_1 -i example/query.fasta -p monomer_ptm -f and i got the .pkl file and msas folder, that was on the featurization CPU part. Now I wanna do the GPU part so will this work? Or something else is required? run_alphafold.sh -d $ALPHAFOLDDB -o test_parafold -m model_1 -i example/query.fasta -p monomer_ptm -t 2021-12-10 -l

Sorry if I'm asking something obvious, just trying to understand how this works.

Zuricho commented 2 years ago

you don't need to use -l, others were right

hrzolix commented 2 years ago

Thank you for your quick response