AlphaFold 2.3.1-foss-2022a-CUDA-11.7.0

AlphaFold can predict protein structures with atomic accuracy even where no similar structure is known

Accessing AlphaFold 2.3.1-foss-2022a-CUDA-11.7.0

To load the module for AlphaFold 2.3.1-foss-2022a-CUDA-11.7.0 please use this command on the BEAR systems (BlueBEAR, BEARCloud VMs, and CaStLeS VMs):

module load bear-apps/2022a
module load AlphaFold/2.3.1-foss-2022a-CUDA-11.7.0

There is a CPU version of this module: AlphaFold 2.3.1-foss-2022a

BEAR Apps Version

2022a

Architectures

EL8-haswell (GPUs: NVIDIA P100) — EL8-icelake (GPUs: NVIDIA A100, NVIDIA A30)

The listed architectures consist of two part: OS-CPU.

  • BlueBEAR: The OS used on BlueBEAR is represented by EL and there are several different processor (CPU) types available on BlueBEAR. More information about the processor types on BlueBEAR is available on the BlueBEAR Job Submission page.
  • BEAR and CaStLeS Cloud VMs: These VMs can have one of two OSes. Those with access to a BEAR Cloud or CaStLeS VM should check that the listed architectures for an application include the OS of VM being used. The VMs, irrespective of OS, will use the haswell CPU type.

AlphaFold Databases

The AlphaFold databases, include both the full and reduced versions, are available in ${BB_APPS_DATA}/AlphaFold. Inside that folder you will find folder(s) dated by download date. Each dated folder contains all of the databases that can be downloaded with the download_all_data.sh script and was downloaded on the date as indicated by the folder name.

Running AlphaFold

For convenience a symbolic link named alphafold points to the run_alphafold.py script. So, once the AlphaFold module is loaded, you can just use alphafold instead of run_alphafold.py or python run_alphafold.py.

The run_alphafold.py script has been modified so that it is sufficient to define the $ALPHAFOLD_DATA_DIR environment variable for it to pick up all the data provided in that location. So, adding export ALPHAFOLD_DATA_DIR=${BB_APPS_DATA}/AlphaFold/20211118 to your jobscript will be sufficient for the centrally downloaded databases to be detected.

Extensions

  • chex 0.1.6
  • contextlib2 21.6.0
  • dm-haiku-0.0.9
  • dm-tree-0.1.8
  • docker 6.0.1
  • immutabledict 2.2.3
  • jmp 0.0.4
  • ml_collections 0.1.1
  • PDBFixer 1.8.1
  • tabulate 0.9.0
  • toolz 0.12.0
  • websocket-client-1.5.1

More Information

For more information visit the AlphaFold website.

Dependencies

This version of AlphaFold has a direct dependency on: Biopython/1.79-foss-2022a CUDA/11.7.0 cuDNN/8.4.1.50-CUDA-11.7.0 foss/2022a HH-suite/3.3.0-gompi-2022a HMMER/3.3.2-gompi-2022a jax/0.3.14-foss-2022a-CUDA-11.7.0 Kalign/3.3.5-GCCcore-11.3.0 NCCL/2.12.12-GCCcore-11.3.0-CUDA-11.7.0 OpenMM/7.7.0-foss-2022a Python/3.10.4-GCCcore-11.3.0 PyYAML/6.0-GCCcore-11.3.0 SciPy-bundle/2022.05-foss-2022a TensorFlow/2.11.0-foss-2022a-CUDA-11.7.0 UCX-CUDA/1.12.1-GCCcore-11.3.0-CUDA-11.7.0

Other Versions

These versions of AlphaFold are available on the BEAR systems (BlueBEAR, BEARCloud VMs, and CaStLeS VMs). These will be retained in accordance with our Applications Support and Retention Policy.

Last modified on 4th April 2023