# EBI PDBe MCP for AI Agents AI Agent Connect > EBI PDBe lets you query the Protein Data Bank in Europe to find 3D protein structures, ligand binding sites, and molecular assemblies. It gives your agent the ability to pull structural biology data directly from the PDBe API without you having to manually search through massive databases or download complex coordinate files. ## Overview - **Category:** the-unthinkable - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_bI09khwqGh15yaAARSveb5XxpbscMGTIKjQ4vrCZ/ai-agent-connect - **Tags:** pdb, protein-structure, 3d-structures, structural-biology, drug-discovery, embl-ebi, crystallography ## Description This EBI PDBe MCP connects you directly to the Protein Data Bank in Europe to explore the world's largest repository of 3D macromolecular structures. If you're a structural biologist or a drug discovery scientist, you know the pain of digging through massive databases to find the right PDB entry. You usually have to hunt for titles, resolutions, and experimental methods across multiple tabs, only to realize the structure you found isn't the right one for your study. This Connector removes that friction. Instead of manual searching, you can ask your agent to pull specific data points like ligand binding sites, molecular formulas, and quality scores. It handles the heavy lifting of querying the official PDBe REST API so you can focus on the actual science. By adding this to your toolkit via the Vinkius catalog, you turn your AI into a research assistant that understands the nuances of quaternary structures and biological assemblies. You get to see the primary citations and PubMed IDs without the copy-paste game. Whether you're trying to identify a binding pocket for a new inhibitor or mapping UniProt sequences to 3D residue numbers, this Connector puts the data you need in front of you in seconds. It's about moving from manual data mining to intelligent inquiry, letting you spend your time on rational drug design and protein chemistry instead of navigating complex web interfaces. ## Tools ### get_binding_sites Find ligand binding site residues and interactions for drug discovery and molecular docking. ### get_secondary_structure Get helix, sheet, and coil assignments per residue to understand protein fold topology. ### get_assemblies Get assembly IDs and composition to see if a protein functions as a monomer, dimer, or higher-order complex. ### get_cofactors Retrieve cofactor and prosthetic group annotations like heme or NAD+ for enzyme catalysis research. ### get_experiment Get the specific experimental method details used to determine a structure. ### get_ligand_monomers Get chemical component IDs, names, formulas, and weights for small molecule ligands. ### get_modified_residues Find non-standard amino acids and nucleotides including their parent compound IDs. ### get_molecules Pull entity IDs, molecule types, sequence lengths, and source organisms for chains and polymers. ### get_mutated_residues See the original and mutated residues to understand how a crystallized construct differs from the wild-type. ### get_publications Find the primary citations and PubMed IDs for a specific structural study. ### get_quality_scores Get global quality metrics like R-factors and resolution to see if a structure is reliable. ### get_related_entries Discover alternative conformations or mutants of the same protein cited in the same publications. ### get_residue_listing Get a full inventory of residue names and numbers organized by entity and chain. ### get_summary Get a PDB entry summary including the title, authors, and resolution using a 4-character ID. ### get_uniprot_mapping Get UniProt accessions and start/end position mappings to cross-reference sequence and structure data. ### search_structures Search the PDB using natural language queries for specific proteins, organisms, or resolutions. ## Prompt Examples **Prompt:** ``` What's the resolution and experimental method for the SARS-CoV-2 main protease (6lu7)? ``` **Response:** ``` I've retrieved the details for **6LU7** (SARS-CoV-2 main protease): * **Resolution:** 2.16 Å * **Method:** X-ray Crystallography * **Summary:** The structure contains the protease in complex with an inhibitor, providing key insights for antiviral drug design. The quality metrics indicate reliable electron density and good geometry for modeling. ``` **Prompt:** ``` Find the ligand binding sites for the hemoglobin structure 4hhb. ``` **Response:** ``` I've located the binding site information for **4HHB** (hemoglobin): * **Binding Sites:** I've identified the specific residues that coordinate the heme groups. * **Cofactors:** The structure contains heme (protoporphyrin IX) groups bound to each of the four globin chains. * **Context:** These heme groups are essential for oxygen binding in the hemoglobin complex. ``` **Prompt:** ``` Search for cryo-EM structures of ribosome complexes from human sources. ``` **Response:** ``` I found several human ribosome structures determined by **cryo-electron microscopy**: | PDB ID | Title | Resolution | Method | | :--- | :--- | :--- | :--- | | [ID 1] | Human 80S ribosome complex | 2.8 Å | Cryo-EM | | [ID 2] | Human ribosome with mRNA | 3.2 Å | Cryo-EM | | [ID 3] | Human 70S ribosome complex | 3.0 Å | Cryo-EM | These structures include various translation factors and mRNA/tRNA complexes. ``` ## Capabilities ### Pull 3D structure summaries Get titles, authors, and resolution for any PDB entry in one go. ### Identify ligand binding sites Retrieve specific residues and interactions for protein-ligand binding. ### Retrieve molecular entity data Get details on protein chains, nucleic acids, and ligands. ### Check structure quality scores Pull R-factors and resolution to verify structure reliability. ### Map UniProt sequences to PDB residues Connect protein sequence positions to 3D structural data. ### Find associated journal publications Get primary citations and PubMed IDs for structural studies. ### Search for specific protein structures Use natural language to find proteins by organism or method. ## Use Cases ### Identifying binding sites for a new inhibitor A scientist asks for the binding pocket residues of a specific protease to see where a drug might fit. The agent uses get_binding_sites to pull the data. ### Cross-referencing UniProt data A bioinformatician needs to map a specific gene's sequence to the 3D coordinates of the protein. The agent uses get_uniprot_mapping to align them. ### Quality checking a new PDB entry A researcher wants to know if a recently deposited structure has a high enough resolution for reliable modeling. The agent uses get_quality_scores. ### Exploring protein fold topology A student asks for the secondary structure counts of a specific enzyme to understand its shape. The agent uses get_secondary_structure. ## Benefits - Skip the manual search: Use search_structures to find specific proteins by name or organism without browsing the PDBe website. - Speed up drug design: Use get_binding_sites to instantly identify key residues in a binding pocket for molecular docking. - Verify data reliability: Check R-factors and resolution with get_quality_scores to ensure you're working with high-quality structures. - Map sequences to 3D space: Use get_uniprot_mapping to connect UniProt sequence data directly to PDB residue numbers. - Understand complex assemblies: Use get_assemblies to see if a protein works as a monomer, dimer, or higher-order complex. - Find related studies: Use get_related_entries to find mutants or alternative conformations from the same research group. ## How It Works The bottom line is you get direct, programmatic access to the world's largest 3D macromolecular structure repository. 1. Subscribe to the EBI PDBe MCP in the Vinkius marketplace. 2. Connect your AI client to the Connector without needing an API key. 3. Ask your agent to find specific protein structures or binding sites. ## Frequently Asked Questions **Does the EBI PDBe MCP require an API key?** No, the PDBe API is public, so you can start querying data immediately after connecting it to your AI client. **Can I use this to find binding sites for drug design?** Yes, the Connector can retrieve specific binding site residues and interactions for various ligands, which is a core use case for drug discovery. **Does this Connector provide 3D coordinate files?** No, it provides structural summaries, metadata, and residue listings rather than the raw coordinate files for simulation. **Can I search for proteins using natural language?** Yes, you can use the search tool to find entries using descriptions like 'insulin receptor kinase' or specific organism names. **How does this help with UniProt sequences?** It allows you to map UniProt accessions to specific PDB residue numbers for easier cross-referencing between sequence and 3D data. **Can I check the quality of a specific PDB entry?** Yes, you can pull global quality metrics like R-factors and resolution to see if a structure is reliable for your research. **Is this Connector good for identifying protein assemblies?** Yes, it can tell you if a protein functions as a monomer, dimer, or a more complex assembly by pulling quaternary structure data. **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.