openfold3
Structure prediction using OpenFold3, an open-weights PyTorch reproduction of AlphaFold3 from the AlQuraishi Lab. Use this skill when predicting protein/nucleic-acid/ligand complex structures with an Apache-2.0-licensed AF3 reimplementation.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-14
- Stars
- ★ 3
- Forks
- 0
01 前言 Front Matter
- name
- openfold3
- license
- Apache-2.0
- category
- biomodels
- metadata
- { "third_party": [ { "kind": "weights", "name": "OpenFold3", "license": "Apache-2.0", "provider": "OpenFold Consortium", "terms_url": "https://github.com/aqlaboratory/openfold-3/blob/main/LICENSE" }, { "kind": "service", "name": "ColabFold MSA server (api.colabfold.com)", "info_url": "https://github.com/sokrypton/ColabFold/wiki", "provider": "Steinegger Lab" } ], "display-name": "OpenFold3" }
- description
- Structure prediction using OpenFold3, an open-weights PyTorch reproduction of AlphaFold3 from the AlQuraishi Lab. Use this skill when predicting protein/nucleic-acid/ligand complex structures with an Apache-2.0-licensed AF3 reimplementation.
- requirements
- [ "gpu" ]
03 SKILL.md 正文
OpenFold3 Structure Prediction
Prerequisites
| Requirement | Minimum | Recommended |
|---|---|---|
| Python | 3.10+ | 3.11 |
| CUDA | 12.1+ | 12.4+ |
| GPU VRAM | 24GB | 80GB (H100) |
| RAM | 32GB | 64GB |
| Disk (weights) | 3GB | - |
How to run
Installation
pip install 'openfold3[cuequivariance]==0.4.1'
The default attention kernel is DeepSpeed DS4Sci_EvoformerAttention. If
DeepSpeed is unavailable, switch to the cuEquivariance triangle kernels (no
build-from-source) by overriding the eval memory settings in
model_config.py (use_deepspeed_evo_attention: False,
use_cueq_triangle_kernels: True). Some pre-built environments already ship
this override; check before re-patching.
Weights
Apache-2.0, ~2.3 GB from HF OpenFold/OpenFold3. The repo is gated (auto-approval) — accept the access form on the HF model page and authenticate (huggingface-cli login or HF_TOKEN) before downloading:
export OPENFOLD_CACHE=~/.openfold3
huggingface-cli download OpenFold/OpenFold3 checkpoints/of3-p2-155k.pt \
--local-dir "$OPENFOLD_CACHE"
run_openfold will also auto-download to $OPENFOLD_CACHE on first run if
egress is open and HF credentials are available (either HF_TOKEN or a prior
huggingface-cli login) with repo access granted. The interactive
setup_openfold helper exists but prompts on stdin; prefer the explicit
download above for non-interactive runs.
Running
export OPENFOLD_CACHE=/path/to/cache
run_openfold predict \
--query_json=queries.json \
--output-dir out/ \
--use-msa-server false \
--use-templates false
run_openfold discovers the checkpoint under $OPENFOLD_CACHE automatically.
Only pass --inference-ckpt-path <file.pt> if you have a non-standard layout
or multiple checkpoints and need to pin one explicitly.
For MSA + templates (slower, higher accuracy), drop the two false flags. The
MSA server is api.colabfold.com; template chain-ID remap hits
data.rcsb.org (GraphQL) — both must be reachable.
Query JSON format
OpenFold3 does not read FASTA. Queries are a JSON object validated by
InferenceQuerySet (pydantic, extra: forbid — unknown keys reject):
{
"queries": {
"my_complex": {
"chains": [
{"molecule_type": "protein", "chain_ids": ["A"], "sequence": "MQIFVK…"},
{"molecule_type": "protein", "chain_ids": ["B", "C"], "sequence": "MVLSPA…"},
{"molecule_type": "ligand", "chain_ids": ["L"], "smiles": "CC(=O)Oc1ccccc1C(=O)O"}
],
"use_msas": true
}
},
"seeds": [42]
}
molecule_type |
required field |
|---|---|
protein / dna / rna |
sequence |
ligand |
smiles or ccd_codes: ["HEM"] |
chain_ids is a list — repeat the same sequence across multiple chain IDs
for homo-oligomers. Per-chain paired_msa_file_paths / main_msa_file_paths
let you supply your own a3m instead of the server.
Key parameters
| Flag | Default | Description |
|---|---|---|
--num-diffusion-samples |
5 | Structures per (query, seed) |
--num-model-seeds |
1 | Number of model seeds per query (multiplies output count alongside JSON seeds and diffusion samples) |
--use-msa-server |
true | ColabFold MMseqs2 server for MSA |
--use-templates |
true | ColabFold template search + RCSB remap |
--inference-ckpt-path |
auto-discovered under $OPENFOLD_CACHE |
Override only — for non-standard layouts or to pin a specific checkpoint file |
Output format
out/
├── summary.txt
├── model_config.json / experiment_config.json
├── inference_query_set.json
└── <query_name>/seed_<N>/
├── <query>_seed_<N>_sample_<k>_model.cif
├── <query>_seed_<N>_sample_<k>_confidences.json # full PAE/pLDDT
├── <query>_seed_<N>_sample_<k>_confidences_aggregated.json
└── timing.json
*_confidences_aggregated.json is the small one to read first:
{
"avg_plddt": 78.96, "ptm": 0.667, "iptm": 0.0, "gpde": 0.73,
"has_clash": 0.0, "sample_ranking_score": 0.133,
"chain_ptm": {"A": 0.667}, "chain_pair_iptm": {}
}
What good output looks like
summary.txtshowsSuccessful Queries: Nmatching your input count- avg_plddt > 70 (single-seq) / > 80 (with MSA)
- ptm > 0.6; for complexes, iptm > 0.5
has_clash: 0.0.cif~50-150 KB per sample for a small protein
Verify
grep -E 'Successful|Failed' out/summary.txt
find out -name '*_model.cif' | wc -l # = queries x json_seeds x num-model-seeds x num-diffusion-samples
Troubleshooting
| Error | Cause | Fix |
|---|---|---|
_deepspeed_evo_attn requires that DeepSpeed be installed |
default eval kernel is DS4Sci on CUDA | install deepspeed (needs nvcc + CUTLASS), or in model_config.py eval block set use_deepspeed_evo_attention: False + use_cueq_triangle_kernels: True (cuEq path; no build) |
CUTLASS_PATH ... not set ... cutlass_library is not installed |
cuEq path still needs the python cutlass_library shim |
pip install nvidia-cutlass |
libXrender.so.1: cannot open shared object file |
rdkit (via pdbeccdutils) needs X11 render libs | apt-get install libxrender1 libxext6 libsm6 |
ModuleNotFoundError: boto3 (or awscrt) |
openfold3.core.data.io.s3 is eager-imported even when weights are local |
pip install boto3 awscrt |
ValidationError: queries / Field required or Input should be an object |
wrong JSON shape | top-level is {"queries": {"<name>": {...}}} (a dict, not a list) |
ValidationError ... settings / Extra inputs are not permitted |
tried to override model config via --runner-yaml |
--runner-yaml is InferenceExperimentConfig only; kernel/memory settings live in model_config.py |
Failed to fetch chain ID mappings from RCSB for N entries |
data.rcsb.org unreachable (allowlist/offline) |
run with --use-templates false, or open egress to data.rcsb.org |
CUDA out of memory |
large complex / many samples | reduce --num-diffusion-samples; the low_mem preset (model_setting_presets.yml) offloads more aggressively |
04 同仓库更多 Skills
workflow-skill-creator
Distills a completed user workflow or interaction into a reusable agent skill. Use when the user asks to turn their workflow, interaction, or multi-step process into a skill, or when they say "make this a skill", "create a skill from what we just did", "package this workflow" or similar. Do not use for creating skills from scratch without an existing workflow (use a generic skill-creator for that).
uv
Checks whether the uv Python package manager is installed and installs it if missing. Ensures uv is on PATH. Use when another skill requires uv as a prerequisite.
uniprot-database
Access protein metadata, function, taxonomy, and sequences across UniProtKB, UniParc, and UniRef. Use when searching for proteins, mapping identifiers, or retrieving functional annotations and publications. Don't use for sequence alignment, protein folding, or sequence similarity search (use specialized skills for those tasks).
unibind-database
Queries the UniBind database for experimentally validated transcription factor (TF) binding sites. Use when retrieving direct TF-DNA interaction datasets, downloading binding site coordinates (BED/FASTA) for local analysis, or listing available datasets by species, cell line, or TF name. Don't use to query specific intervals, locations, genes, motif models or expression data.
ucsc-conservation-and-tfbs
Fetch Evolutionary Conservation scores (phyloP, phastCons) and Transcription Factor Binding Sites (TFBS) from the UCSC Genome Browser. Use when analyzing whether genomic variants or regions are evolutionarily conserved, functionally important, or bounded by TF regulators across major projects (ENCODE, JASPAR, ReMap).
string-database
Query the STRING database for protein-protein interactions (PPIs), functional enrichment, and homology. Use when the user asks about interactions between specific proteins, interaction evidence, confidence scores, protein interaction partners, or pathway enrichments.