Msa Structure Prediction Pipeline

What it does

When your agent uses it

• The user wants to predict a protein structure with maximum accuracy using MSA context
• Run the full AlphaFold3-style pipeline
• Generate MSA-informed structure predictions
• Improve structure prediction accuracy by providing evolutionary information

Example prompts

• “/msa-structure-prediction-pipeline”

Requirements

• Python 3
• Docker
• A credential in NGC_API_KEY
• Pre-approved tools (allowed-tools): Bash, Read, Write, AskUserQuestion

Workflow steps

2 steps, taken from the step headings in SKILL.md.

1. 1Search for MSA with MSA-Search
2. 2Predict structure with OpenFold3

What it can do on your machine

Read from SKILL.md and the folder at commit dfdd080. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Bash
  • Read
  • Write
  • AskUserQuestion

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are python).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Hosts in commands or code, which the agent is likely to contact:

  • health.api.nvidia.com

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names these keys or tokens, usually read from environment variables:

  • NGC_API_KEY

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Msa Structure Prediction Pipeline loads about 1.6k tokens when it runs. Until then it costs about 252 tokens; SKILL.md has 287 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~252

When it runs · the whole SKILL.md, loaded when a task matches

~1.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

• NotePre-approves every shell command (allowed-tools: Bash)SKILL.md

  allowed-tools: Bash, Read, Write, AskUserQuestion

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from NVIDIA/skills at commit dfdd080, republished under its Apache-2.0 licence (© NVIDIA). 287 words, ~1,640 tokens.

name

msa-structure-prediction-pipeline

description

NOTE: your protein sequence and the retrieved MSA alignment are transmitted to external NVIDIA-hosted APIs (health.api.nvidia.com) on every call. Use local NIM containers for confidential or proprietary sequences. Run a complete protein structure prediction pipeline using NVIDIA BioNeMo NIMs: search for MSA alignments with MSA-Search (ColabFold), then predict the structure with OpenFold3 using the retrieved alignments. Use this skill whenever the user wants to predict a protein structure with maximum accuracy using MSA context, run the full AlphaFold3-style pipeline, generate MSA-informed structure predictions, or improve structure prediction accuracy by providing evolutionary information. Triggers on: MSA structure prediction pipeline, structure prediction pipeline, MSA-informed prediction, OpenFold3, ColabFold MSA, AlphaFold3 pipeline, protein structure, homology search, a3m alignment, UniRef30, NIM microservice. This pipeline chains MSA-Search and OpenFold3.

allowed-tools

Bash, Read, Write, AskUserQuestion

license

Apache-2.0 AND CC-BY-4.0

Step 1: MSA-Search  →  Step 2: OpenFold3
(Search homologs)       (Predict structure with MSA)

import requests, json, os
from pathlib import Path

NGC_API_KEY = os.getenv("NGC_API_KEY")
HOSTED = True

query_sequence = "<YOUR_PROTEIN_SEQUENCE>"

if HOSTED:
    msa_url = "https://health.api.nvidia.com/v1/biology/colabfold/msa-search/predict"
    headers = {"Content-Type": "application/json",
               "Authorization": f"Bearer {NGC_API_KEY}"}
else:
    msa_url = "http://localhost:8000/biology/colabfold/msa-search/predict"
    headers = {"Content-Type": "application/json"}

payload = {
    "sequence": query_sequence,
    "databases": ["Uniref30_2302", "colabfold_envdb_202108"],
    "e_value": 0.0001,
    "output_alignment_formats": ["a3m"],
}

r = requests.post(msa_url, headers=headers, json=payload)
r.raise_for_status()
msa_result = r.json()

# Extract the A3M alignment
a3m_alignment = msa_result["alignments"]["Uniref30_2302"]["a3m"]["alignment"]

# Save for reference
with open("query_msa.a3m", "w") as f:
    f.write(a3m_alignment)

# Count sequences in alignment
n_seqs = a3m_alignment.count(">")
print(f"Step 1 complete: found {n_seqs} homologous sequences")
print(f"MSA saved to query_msa.a3m")

if HOSTED:
    of3_url = "https://health.api.nvidia.com/v1/biology/openfold/openfold3/predict"
else:
    of3_url = "http://localhost:8000/biology/openfold/openfold3/predict"

# Build the OpenFold3 MSA structure from the retrieved alignment
msa_data = {
    "uniref30": {
        "a3m": {
            "alignment": a3m_alignment,
            "format": "a3m"
        }
    }
}

# Optionally also include colabfold_envdb alignment if requested
# env_alignment = msa_result["alignments"]["colabfold_envdb"]["a3m"]["alignment"]
# msa_data["colabfold_env"] = {"a3m": {"alignment": env_alignment, "format": "a3m"}}

payload = {
    "inputs": [{
        "input_id": "prediction_with_msa",
        "output_format": "pdb",
        "molecules": [
            {
                "type": "protein",
                "sequence": query_sequence,
                "diffusion_samples": 1,
                "msa": msa_data
            }
        ]
    }]
}

r = requests.post(of3_url, headers=headers, json=payload, timeout=300)
r.raise_for_status()
result = r.json()

output = result["outputs"][0]
for i, sample in enumerate(output["structures_with_scores"]):
    fmt = sample["format"]
    filename = f"predicted_structure_{i+1}.{fmt}"
    with open(filename, "w") as f:
        f.write(sample["structure"])
    print(f"\nStep 2 complete: {filename} saved")
    print(f"  Confidence:  {sample['confidence_score']:.4f}")
    print(f"  pLDDT:       {sample['complex_plddt_score']:.4f}")
    print(f"  pTM:         {sample['ptm_score']:.4f}")

# Minimal MSA (single sequence — same as no MSA context):
minimal_msa = {
    "main": {
        "a3m": {
            "alignment": f">query\n{query_sequence}",
            "format": "a3m"
        }
    }
}

# Paired MSA search endpoint for complexes:
msa_paired_url = "https://health.api.nvidia.com/v1/biology/colabfold/msa-search/paired/predict"
paired_payload = {
    "sequences": [chain_A_sequence, chain_B_sequence],
    "e_value": 0.0001,
}

© NVIDIA, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 4 other files in skills/bionemo-msa-structure-prediction-pipeline of NVIDIA/skills.

• SKILL.md
• BENCHMARK.md
• evals/evals.json
• skill-card.md
• skill.oms.sig

Open the folder on GitHubat commit dfdd080

Used in 1 other repository

We found 2 copies of this SKILL.md (exact, near-identical or edited) in other folders, from 1 other GitHub owner. This page covers the copy in NVIDIA/skills, which our catalogue first saw on October 7, 2026.

• NVIDIA-BioNeMo/bionemo-agent-toolkitnim-skills/meta-skills/msa-structure-prediction-pipelineskills/bionemo-agent-toolkit/skills/msa-structure-prediction-pipeline