EBI PDBe MCP for AI. Analyze 3D Protein Structures and Interactions.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Connect to your AI in seconds.
EBI PDBe provides immediate access to 3D protein structures, ligand interactions, and molecular assemblies from the Protein Data Bank in Europe.
This MCP lets you analyze complex biological systems—from determining if a protein forms a dimer to pinpointing exact binding sites for drug design—all without downloading massive coordinate files.
What your AI can do
Get assemblies
Determines the biological assembly state of a protein to see if it's a monomer, dimer, or larger complex.
Get binding sites
Locates specific residues and interactions that form binding pockets for small molecules.
Get cofactors
Lists essential cofactors, like metal ions or heme groups, necessary for the protein's function.
Identifies if a protein functions as a monomer, dimer, or higher-order complex by analyzing its quaternary structure.
Pinpoints exact residues where small molecule ligands bind to proteins, which is critical for drug design research.
Retrieves global quality metrics like resolution and R-factors, letting you immediately vet a structure's scientific validity.
Maps residue numbers between general protein sequence databases (UniProt) and the 3D structural annotations (PDB).
Performs full-text searches across thousands of structures using natural language queries, finding candidates based on method or organism.
Ask an AI about this
Waiting for input…
EBI PDBe: 16 Analysis Tools
Use these specialized tools to perform deep analyses on molecular assemblies, ligand interactions, cofactors, and experimental data from the Protein Data Bank in Europe.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using EBI PDBe on VinkiusGet Assemblies
Determines the biological assembly state of a protein to see if it's a monomer, dimer, or larger complex.
Get Binding Sites
Locates specific residues and interactions that form binding pockets for small...
Get Cofactors
Lists essential cofactors, like metal ions or heme groups, necessary for the...
Get Experiment
Retrieves detailed information about the experimental method used to determine the...
Get Ligand Monomers
Gets a list of small molecule ligands bound in the crystal structure, including...
Get Modified Residues
Shows non-standard amino acids or nucleotides present in the sequence data.
Get Molecules
Retrieves detailed inventory of all molecular entities, like chains and polymers, within a given structure.
Get Mutated Residues
Highlights specific residues that have been engineered or mutated compared to the...
Get Publications
Finds primary citations and PubMed IDs associated with a specific structural...
Get Quality Scores
Calculates global metrics for the structure, giving an immediate assessment of its...
Get Related Entries
Discovers alternative conformations or mutant versions of a protein that have been...
Get Residue Listing
Generates an inventory list of residues, organized by chain and entity, for detailed inspection.
Get Secondary Structure
Assigns the protein's fold topology by counting helices, strands, and coils per residue.
Get Summary
Provides a quick overview of the PDB entry, including its title, authors, and...
Get Uniprot Mapping
Creates a cross-reference map linking UniProt sequence positions to specific...
Search Structures
Searches across the entire database using natural language queries for structures...
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token
from cloud.vinkius.com. For OAuth-protected servers, expand
Advanced settings to add credentials.
Start a conversation
Open a new chat. The EBI PDBe integration is available immediately — no restart needed.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with EBI PDBe, then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by PDBe (Protein Data Bank in Europe). All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This connection provides 16 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Checking protein structure data today means jumping between five different databases and writing specific API queries every time.
If you're working manually, finding a single binding site requires several steps. You might pull up the PDB ID in one window to get the summary; then open another tab to check for associated ligands. If you want to know if that structure is reliable enough for publication, you have to run a separate quality assessment query and cross-reference the authors' papers manually.
With this MCP connected via Vinkius, your agent handles all that coordination. You simply ask: 'Tell me about the ligand interactions in 6lu7.' The system pulls together the binding sites, checks the cofactors, and even gives you the quality score—all in one conversation.
You get immediate access to structural details using `get_binding_sites`.
Previously, identifying a pocket meant downloading coordinate files, manually viewing them, and hoping the accompanying documentation covered everything. If you missed checking for cofactors or mutations, your entire hypothesis could fail without you realizing it.
Now, querying `get_binding_sites` gives you precise residue information on where ligands dock, while tools like `get_cofactors` tell you what other essential molecules are involved. It’s a complete structural picture that saves hours of manual file handling.
What your AI can actually do with this
This connector lets your AI client function like a structural biology research assistant, giving you direct access to the world’s repository of experimentally determined macromolecular structures. You can query specific proteins or even search across entire classes of molecules using natural language. For example, you don't need to know the exact PDB ID; you just ask for 'SARS-CoV-2 spike protein,' and the tool finds it.
From there, you can drill down into functional details: identifying which small molecule ligands are bound, checking if a structure is reliable enough for publication, or mapping out how different protein chains assemble together. It’s about getting immediate, actionable structural data without having to write complex API calls or download large files first.
You manage all this complexity through the Vinkius catalog, making it simple to plug into your existing workflow.
019dea5f-351a-73b1-9107-2a979b0b2bd6 
Here's how it actually works
The bottom line is: you get deep, specific structural knowledge without ever touching a database console or worrying about endpoint details.
Subscribe to this MCP and connect your preferred AI client. No API key is needed because the underlying PDBe data is public.
Ask your agent a structural question, like 'What cofactors are bound in 4hhb?' or 'Show me all complexes related to X.'
The MCP executes the query against the official PDBe REST API and returns structured biological data directly to your client for analysis.
Who is this actually for?
This is for computational biologists and drug discovery scientists who spend their time correlating sequence data with physical structure. You're the one staring at millions of coordinates, needing to know if a binding pocket exists before you can even start designing a molecule.
Needs to quickly retrieve metadata and quality scores for dozens of candidate proteins to determine which ones are reliable enough for publication.
Must identify specific ligand binding sites and analyze cofactors to guide the rational design of new inhibitors or drugs.
Needs to map sequence information from general databases onto 3D structural residues, ensuring data consistency across different sources.
What Changes When You Connect
You can instantly assess the structural reliability of any candidate using get_quality_scores, saving time spent on manual data vetting. Knowing if a structure is high-resolution or poor geometry changes your entire hypothesis.
Drug development gets an edge when you use get_binding_sites to pinpoint exactly where potential ligands fit and interact, guiding rational drug design efforts.
When validating sequence data, the get_uniprot_mapping tool eliminates guesswork by creating a reliable cross-reference between general protein databases and 3D residue coordinates.
If you're studying enzyme function, checking for essential metal ions or cofactors using get_cofactors is critical, letting you understand what makes the protein work in vivo.
The search_structures tool lets you skip manual browsing. You just ask for 'cryo-EM structures of ribosome complexes,' and it pulls up all candidates immediately.
See it in action
Evaluating a Novel Inhibitor
A chemist needs to know if their new molecule can bind to the target protein. They use search_structures first, then call get_binding_sites on promising candidates to see available pockets; finally, they check get_ligand_monomers to confirm the type of small molecules that fit.
Investigating Protein Evolution
A bioinformatician is comparing two related proteins. They use get_related_entries to find alternative conformations and then call get_mutated_residues to see exactly how the sequence differs from the wild-type.
Determining Protein Architecture
A structural biologist is trying to understand if a protein acts alone or in groups. They use get_assemblies to confirm its quaternary structure, then call get_secondary_structure to map out the specific helices and sheets involved.
The honest tradeoffs
Assuming full sequence data
Just looking at a PDB ID summary without checking for mutations. You might assume the structure is perfect, but it's based on a different variant.
Always run get_mutated_residues and get_uniprot_mapping to verify that the coordinates you are reading match the specific sequence variants you care about.
Ignoring function details
Searching for a general protein type without checking if it needs metal ions. You waste time designing an inhibitor for a cofactor-dependent enzyme.
Before proceeding, check get_cofactors to ensure the necessary cofactors are annotated in the structure.
Overlooking assembly state
Treating a complex as if it were one single protein. This misses crucial information about which subunit is doing what.
Start by running get_assemblies. This immediately tells you the correct context (monomer, dimer, etc.) before you analyze any specific binding sites.
When It Fits, When It Doesn't
Use this MCP if your core question revolves around physical geometry or molecular interaction. Specifically, use it when you need to know where a ligand binds (get_binding_sites), how the protein is built from subunits (get_assemblies), or what its overall quality is (get_quality_scores). Don't use it if your problem is purely textual; for example, if you just need to compare gene names across species without structural context, a general sequence alignment tool is better. If you are trying to find related structures, start with search_structures before diving into detailed analyses like get_residue_listing. This MCP is for the 'how'—the physical reality of the molecule.
Questions you might have
How do I use the get_binding_sites tool? +
You call get_binding_sites with a PDB ID to retrieve specific residues and interactions that form binding pockets, which is essential for drug discovery.
What does search_structures do? +
search_structures allows you to query the entire database using natural language—for example, 'cryo-EM structures of ribosome complexes'—to find relevant PDB IDs immediately.
Should I use get_assemblies or get_molecules? +
get_assemblies gives you the high-level context (dimer vs. monomer), while get_molecules provides a detailed list of every type and chain present in the structure.
Is it possible to map UniProt data with this MCP? +
Yes, you use get_uniprot_mapping to generate a cross-reference table that links sequence positions from general databases (UniProt) directly to the residue numbers in the 3D structure.
How can I verify the reliability of a protein structure using get_quality_scores? +
It returns global metrics like R-factors and resolution. This lets you assess how reliable the structure is, which structural biologists check first before drawing any conclusions.
If I need to identify all distinct components in a complex, should I use get_molecules? +
Yes, this tool returns IDs and types for every molecular entity. You get chain assignments, sequence lengths, weights, and source organisms listed across the whole structure.
Where can I find alternative forms or related structures of a known protein using get_related_entries? +
This tool finds other PDB entries that are structurally linked to your original query. It’s useful for comparing different conformations, mutants, or complexes.
How do I find essential metal ions or prosthetic groups using get_cofactors? +
It retrieves annotations for cofactors like heme, NAD+, and various metal ions. This tells you exactly which non-protein chemical components are bound to the structure.
Do I need an API key? +
No. The PDBe API is completely public and requires no authentication. Enter any placeholder value in the API key field to activate the server immediately.
What types of structures are available? +
The PDBe contains over 200,000 experimentally determined 3D structures of proteins, nucleic acids, and complex assemblies. Structures are determined by X-ray crystallography, cryo-electron microscopy (cryo-EM), NMR spectroscopy, and other methods. This includes enzymes, receptors, antibodies, viral proteins, ribosomes, and drug-target complexes.
Can I find drug binding sites? +
Yes. Use get_binding_sites to retrieve all annotated ligand binding pockets with their constituent residues. Combine with get_ligand_monomers to identify the small molecules bound in the structure, and get_cofactors for prosthetic groups. This workflow is essential for structure-based drug design and virtual screening target preparation.
We've already built the connector for EBI PDBe. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 16 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.