Installing SBGrid Software
Using the SBGrid Environment
Support for Site Administrators
Hardware Support Notes
Getting Help
Support for Developers
CUDA SDK is required
Alphapulldown requires that the CUDA Toolkit SDK be installed. We use version 11.1 and have had success with latest versions. The executables provided in this package are not redistributable so we do not include them with our CUDA libraries.
Required Databases and Parameters AlphaPullDown requires the AlphaFold set of (large) genetic databases that must be downloaded separately. See https://github.com/deepmind/alphafold#genetic-databases for more information.
These databases can be downloaded with the download script included in AlphaFold and the aria2c program, both of which are available in the SBGrid collection. Note that these databases are large in size (> 2Tb) and may require a significant amount of time to download.
/programs/x86_64-linux/alphafold/2.3.2/alphafold/scripts/download_all_data.sh <destination path>
*AlphaPullDown 0.37.0 *
#!/usr/bin/env bash
## SBGrid AlphaPullDown 0.30.7 example
## from https://github.com/KosinskiLab/AlphaPulldown/blob/main/example_1.md
## Args: none
##
## June 27, 2023
:<<ACKNOWLEDGE
If your use of SBGrid compiled software was an important element in
your publication, please include the following statement in your work:
"Software used in the project was installed and configured by SBGrid
(cite: eLife 2013;2:e01456, Collaboration gets the most out of software.)
ACKNOWLEDGE
# Set these to local requirements:
module load cuda/11.2
export LD_LIBRARY_PATH=/programs/x86_64-linux/cuda/11.2/lib64
export MY_DATADIR=/programs/local/alphafold-2.3.2
export SBGRID_ALLOW=true # sbgrid devs only
source /programs/sbgrid.shrc
export ALPHAPULLDOWN_X=0.30.7
export MY_AP_DIR="./AlphaPulldown/example_data"
export MY_OUTDIR=outdir
mkdir -p ${MY_OUTDIR}
# Clone github repo in order to get example data here:
git clone https://github.com/KosinskiLab/AlphaPulldown.git
# Step 1
create_individual_features.py \
--fasta_paths=${MY_AP_DIR}/baits.fasta,${MY_AP_DIR}/example_1_sequences.fasta \
--data_dir=${MY_DATADIR} \
--save_msa_files=False \
--output_dir=${MY_OUTDIR} \
--use_precomputed_msas=False \
--max_template_date="2023-01-01" \
--skip_existing=False \
# Step 2
run_multimer_jobs.py --mode=pulldown \
--num_cycle=1 \
--num_predictions_per_model=1 \
--output_path=${MY_OUTDIR} \
--data_dir=${MY_DATADIR} \
--protein_lists=${MY_AP_DIR}/baits.txt,${MY_AP_DIR}/candidates.txt \
--monomer_objects_dir=${MY_OUTDIR} \
--job_index=2