Predict joint biomolecular structures for proteins, DNA/RNA, ligands, and ions.
npx skills add https://github.com/yjcyxky/biominer-skills --skill alphafold3
Please help me install this Agent Skill. Skill: alphafold3 Source: https://github.com/yjcyxky/biominer-skills/tree/main/alphafold3 Command: npx skills add https://github.com/yjcyxky/biominer-skills --skill alphafold3
This Skill enables automated joint prediction of biomolecular structures for proteins, DNA/RNA, ligands, ions, and their complexes, reducing manual modeling time and enabling rapid hypothesis testing in structural biology.
Run AlphaFold3 in a GPU-enabled Docker environment using the official image, mounting model directories, genetic databases, and input/output folders. Example:
docker run --rm --gpus all
-v /path/to/models:/root/models
-v /path/to/databases:/root/public_databases
-v $(pwd)/input:/root/af_input
-v $(pwd)/output:/root/af_output
ghcr.io/open-prophetdb/alphafold3:arm64-blackwell
python run_alphafold.py --json_path=/root/af_input/fold_input.json --model_dir=/root/models --output_dir=/root/af_output
High-intent search queries and answers about installing and using this skill.
You can predict protein-DNA/RNA complex structures by defining multiple entity types within a single formatted input JSON file, allowing AlphaFold3 to jointly model the multi-component assembly and their interactions in one run.
Yes, you must obtain licensed AlphaFold3 model parameters from Google and download genetic databases for MSA searches before running joint biomolecular structure predictions in the Docker environment.
Yes, AlphaFold3 supports drug discovery and protein engineering workflows by providing diffusion-based, high-accuracy structure prediction for biomolecular assemblies including protein-ligand interactions and complex formations.
Running AlphaFold3 requires a Docker-enabled GPU environment, licensed model parameters, genetic databases for MSA searches, and correctly formatted input JSON files to execute the automated structure prediction workflows.
AlphaFold3 handles joint prediction of multi-component assemblies by using diffusion-based modeling to simultaneously predict structures for proteins, DNA/RNA, ligands, and ions defined in the input JSON, reducing manual modeling time.
PKU-YuanGroupPredict protein structures for monomers and multimers with AlphaFold2.
LeonChaoXRetrieve AlphaFold DB protein structures and confidence metrics by UniProt accession.
PKU-YuanGroupPredict protein, nucleic-acid, and small-molecule complex structures with Boltz-2.
sanand0Predict 3D binding poses of small molecules to protein targets using diffusion models.
