Predict structures for designed protein sequences and protein complexes with AlphaFold2 or ColabFold. Inspect structure files and confidence scores to assess folding and chain interfaces, with guidance for local runs, Modal execution, and common prediction errors.
中文简介正在准备中。当前显示英语简介。
Validate protein designs using AlphaFold2 structure prediction. Use this skill when: (1) Validating designed sequences fold correctly, (2) Predicting binder-target complex structures, (3) Calculating confidence metrics (pLDDT, pTM, ipTM), (4) Self-consistency validation of designs, (5) Multi-chain complex prediction with AlphaFold-Multimer. For faster single-chain prediction, use esm. For QC thresholds, use protein-qc.
实测评分 未实测
查看安装方法安装前可以查看交给 AI 代理的指令内容。
| Requirement | Minimum | Recommended |
|---|---|---|
| Python | 3.8+ | 3.10 |
| CUDA | 11.0+ | 12.0+ |
| GPU VRAM | 32GB | 40GB (A100) |
| RAM | 32GB | 64GB |
| Disk | 100GB | 500GB (for databases) |
First time? See Installation Guide to set up Modal and biomodals.
cd biomodals
modal run modal_colabfold.py \
--input-faa sequences.fasta \
--out-dir output/
GPU: A100 (40GB) | Timeout: 3600s default
git clone https://github.com/deepmind/alphafold.git
cd alphafold
python run_alphafold.py \
--fasta_paths=query.fasta \
--output_dir=output/ \
--model_preset=monomer \
--max_template_date=2026-01-01
modal run modal_esmfold.py \
--sequence "MKTAYIAKQRQISFVK..."
| Parameter | Default | Options | Description |
|---|---|---|---|
--model_preset | monomer | monomer/multimer | Model type |
--num_recycle | 3 | 1-20 | Recycling iterations |
--max_template_date | - | YYYY-MM-DD | Template cutoff |
--use_templates | True | True/False | Use template search |
output/
├── ranked_0.pdb # Best model
├── ranked_1.pdb # Second best
├── ranking_debug.json # Confidence scores
├── result_model_1.pkl # Full results
├── msas/ # MSA files
└── features.pkl # Input features
import pickle
with open('result_model_1.pkl', 'rb') as f:
result = pickle.load(f)
plddt = result['plddt']
ptm = result['ptm']
iptm = result.get('iptm', None) # Multimer only
pae = result['predicted_aligned_error']
$ python run_alphafold.py --fasta_paths complex.fasta --model_preset multimer
[INFO] Running MSA search...
[INFO] Running model 1/5...
[INFO] Running model 5/5...
[INFO] Relaxing structures...
Results:
ranked_0.pdb:
pLDDT: 87.3 (mean)
pTM: 0.78
ipTM: 0.62
PAE (interface): 8.5
Saved to output/
What good output looks like:
Should I use AlphaFold?
│
├─ What are you predicting?
│ ├─ Single protein → ESMFold (faster)
│ ├─ Protein-protein complex → AlphaFold/ColabFold ✓
│ ├─ Protein + ligand → Chai or Boltz
│ └─ Batch of sequences → ColabFold ✓
│
├─ What do you need?
│ ├─ Highest accuracy → AlphaFold/ColabFold ✓
│ ├─ Fast screening → ESMFold
│ └─ MSA-free prediction → Chai or ESMFold
│
└─ Which AF2 option?
├─ Local installation → Full control, slow setup
├─ ColabFold → Easier, MSA server
└─ Modal → Recommended for batch
| Campaign Size | Time (A100) | Cost (Modal) | Notes |
|---|---|---|---|
| 100 complexes | 1-2h | ~$8 | With MSA server |
| 500 complexes | 5-10h | ~$40 | Standard campaign |
| 1000 complexes | 10-20h | ~$80 | Large campaign |
Per-complex: ~30-60s with MSA server.
find output -name "ranked_0.pdb" | wc -l # Should match input count
Low pLDDT regions: May indicate disorder or poor design Low ipTM: Interface not confident, check hotspots High PAE off-diagonal: Chains may not interact OOM errors: Use ColabFold with MSA server instead
| Error | Cause | Fix |
|---|---|---|
RuntimeError: CUDA out of memory | Sequence too long | Use A100 or split prediction |
KeyError: 'iptm' | Running monomer on complex | Use multimer preset |
FileNotFoundError: database | Missing MSA databases | Use ColabFold MSA server |
TimeoutError | MSA search slow | Reduce num_recycles |
Next: protein-qc for filtering and ranking.
还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Support protein design with cloud laboratory testing for binding, expression, thermal stability, and enzyme activity. Covers sequence optimization, submitting experiments through the Adaptyv API, checking progress, and downloading experimental results.
中文简介正在准备中。当前显示英语简介。
Build flexible daily plans that account for ADHD-related difficulties with time estimation and getting started. Narrow priorities to three tasks, add transition buffers, and choose small next actions that match the energy available that day.
中文简介正在准备中。当前显示英语简介。
Build Python workflows for data recorded over time, including classification, forecasting, anomaly detection, and pattern search. Using aeon, the skill guides preprocessing, algorithm selection, feature extraction, and evaluation through interfaces compatible with scikit-learn.
中文简介正在准备中。当前显示英语简介。
Use a browser to research topics, extract page information, fill forms, and test websites. Guides agents through identifying page controls, interacting with web apps, saving screenshots or PDFs, and reusing saved login sessions.
中文简介正在准备中。当前显示英语简介。
Combine health measurements, exercise, sleep, nutrition, and mental health records to examine trends, unusual changes, and estimated risks. The skill produces personalized suggestions and text or interactive HTML reports using local data without external API calls.
中文简介正在准备中。当前显示英语简介。
Retrieve predicted protein structures from AlphaFold DB using UniProt identifiers. Download structure files, examine confidence scores that indicate prediction reliability, and summarize multiple proteins or obtain species datasets for larger structural analyses.
中文简介正在准备中。当前显示英语简介。