Skip to content

AlphaFoldΒΆ

AlphaFold is an AI system developed by DeepMind that makes state-of-the-art accurate predictions of a protein's structure from its amino-acid sequence.

AlphaFold is made available on the cluster as a shared Apptainer container image. This should be run on a GPU Compute partition.

You can use the apptainer/alphafold2 module to add a convenient alias: run_alphafold_apptainer, which will run the run_alphafold.py within the container; The alias will also bind-mount the AlphaFold database base path ($AF2DB, which is set to /opt/alphafold_databases/) into the container on /db.

To use the run_alphafold_apptainer alias in a non-interactive/SLURM batch script, add the following in your script before using the alias:

Terminal

shopt -s expand_aliases

See the AlphaFold documentation for usage information. Example:

Terminal

#!/bin/bash
#SBATCH ....
#SBATCH ....
module load apptainer/alphafold2
shopt -s expand_aliases

INPUT=/home/doeja01p/alphafold_test/in
OUTPUT=/home/doeja01p/alphafold_test/out

run_alphafold_apptainer \
--use_gpu_relax \
--fasta_paths=${INPUT}/T1050.fasta \
--output_dir=$OUTPUT \
--max_template_date=2020-05-14 \
--model_preset=monomer_casp14 \
--benchmark \
--data_dir=/db \
--uniref90_database_path=/db/uniref90/uniref90.fasta \
--mgnify_database_path=/db/mgnify/mgy_clusters_2022_05.fa \
--template_mmcif_dir=/db/pdb_mmcif/mmcif_files \
--obsolete_pdbs_path=/db/pdb_mmcif/obsolete.dat \
--bfd_database_path=/db/bfd/bfd_metaclust_clu_complete_id30_c90_final_seq.sorted_opt \
--uniref30_database_path=/db/uniref30/UniRef30_2021_03 \
--pdb70_database_path=/db/pdb70/pdb70