EBI Proteins API MCP for AI. Deep Dive into Protein Structure and Variation Data
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Connect to your AI in seconds.
The EBI Proteins API MCP connects your AI client directly to UniProt, giving you access to millions of protein entries and their full functional data.
You can pull sequences, map genetic variants, find binding sites, and check post-translational modifications all in one query. It's a deep dive into the entire protein biology knowledge base.
What your AI can do
Get antigen
Retrieves peptide regions used for antibody generation, useful when targeting specific protein parts.
Get coordinates
Returns the precise genomic location (chromosome, start/end) and Ensembl IDs for a given protein.
Get protein features
Retrieves detailed annotations about the sequence, including domains, binding sites, active sites, and signal peptides.
Fetch an entire protein entry using its UniProt accession, including names, organism data, and cross-references.
Get annotated details on domains, binding sites, active sites, and transmembrane regions for any given sequence.
Access curated variant data aggregated from multiple large-scale studies, assessing clinical significance and consequence type.
Find the exact chromosome coordinates, Ensembl gene IDs, and transcript mapping for a specific protein.
Query data on post-translational modifications (PTMs) or mass spectrometry peptide evidence for validation.
Ask an AI about this
Waiting for input…
EBI Proteins API: 16 Tools for Bioinformatics
These tools let you query every facet of protein biology—from genomic mapping to post-translational modifications—with precise, programmatic calls.
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 Proteins API on VinkiusGet Antigen
Retrieves peptide regions used for antibody generation, useful when targeting specific protein parts.
Get Coordinates
Returns the precise genomic location (chromosome, start/end) and Ensembl IDs for a...
Get Protein Features
Retrieves detailed annotations about the sequence, including domains, binding sites...
Get Genecentric
Provides a view of how many related proteins exist for a specific gene ID within a...
Get Mutagenesis
Lists known mutagenesis experiments, detailing the wild-type residue, mutant...
Get Protein
Fetches a complete record for any protein using its standard UniProt accession code.
Get Proteome
Gets high-level information about an entire organism's protein set, such as total protein count or taxonomy details.
Get Proteomics
Provides mass spectrometry data to show which peptides were experimentally detected...
Get Proteomics Ptm
Specifies residue-level positions and evidence counts for post-translational...
Get Taxonomy
Looks up scientific names, ranks, and lineage connections using an NCBI taxon ID.
Get Variation
Gathers genetic variant data for a protein from multiple sources, noting clinical...
Search Features By Type
Searches across proteins to find specific types of features like binding sites or transmembrane regions.
Search Proteins
Finds a summarized list of proteins by searching based on gene name, organism, or keyword.
Search Proteomes
Searches for entire proteomes using general terms like 'homo sapiens' or...
Search Taxonomy
Finds correct taxonomy entries by name, providing the necessary IDs to start a...
Search Variation
Searches for clinically relevant genetic variants based on their consequence type or...
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 Proteins API 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 Proteins API, 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 EMBL-EBI Proteins API. 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.
Manually assembling protein data from different databases used to take days.
Right now, if you need a complete picture of a protein, you're clicking through multiple web interfaces. You pull the sequence from UniProt, then jump to ClinVar for variants, and maybe check PDB for structure, all while tracking down the correct accession IDs just to make sure everything lines up.
With this MCP, your agent handles that entire process in one query. Instead of jumping between platforms, you ask for what you need—be it sequence details or variant status—and get a single, compiled answer set.
Get the full picture with the get_protein-powered workflow
The tedious part of gathering coordinates and functional data is eliminated. You don't have to manually cross-reference a protein ID against a genome mapping tool, then separately query its known binding sites.
You just ask for the full picture using tools like get_coordinates and get_protein_features; the answer comes back mapped, annotated, and ready to use.
What your AI can actually do with this
Working with protein data shouldn't mean juggling five different databases. This MCP lets you query the comprehensive UniProt knowledge base for everything related to proteins—from basic sequence retrieval to complex functional annotations. Need to know if a certain variant is clinically significant? You can pull that aggregated information from sources like ClinVar and gnomAD.
Want to map where a protein sits on a genome? Or check out every known binding site and active domain? It all comes together here. By connecting this MCP via Vinkius, your agent acts as an expert molecular biology assistant, giving you direct access to the full range of data sourced by EMBL-EBI.
You get precise details like mass spectrometry evidence for specific modifications or detailed descriptions of mutagenesis experiments. The result is a single source of truth that bypasses hours of manual searching across multiple bioinformatics platforms.
019dea5f-8355-7266-97eb-6c9b62622924 
Here's how it actually works
The bottom line is that you ask a complex biological question once, and the MCP handles all the data sourcing and structuring for you.
First, connect your AI client to this MCP via Vinkius and confirm access.
Next, give your agent a specific biological question—for instance, 'What are the known variants for TP53?' or 'Show me the domain structure of EGFR.'
Finally, your agent executes the necessary queries using the specialized tools and delivers structured data directly to you.
Who is this actually for?
Molecular biologists and bioinformaticians who are tired of cross-referencing protein IDs across five different databases. This is for researchers needing to synthesize deep, multi-source biological evidence quickly.
Needs to find binding sites and characterize specific domains (e.g., using get_protein_features) to guide wet lab experiments.
Requires aggregating variant data from sources like ClinVar or gnomAD (using get_variation) to assess pathogenicity for patient reports.
Needs to programmatically map proteins to genome coordinates and build pipelines using taxonomy IDs or proteome searches.
What Changes When You Connect
You stop guessing which database holds the variant info. By calling get_variation, you pull aggregated data from sources like ClinVar and gnomAD in one step.
Forget manually tracing protein IDs across genomic maps. Use get_coordinates to map a protein directly to its chromosome location (GRCh38) with Ensembl IDs.
Need context for an entire species? Instead of searching piece by piece, use search_proteomes and get_proteome to view the total count and taxonomy status immediately.
Structural analysis is faster than ever. The get_protein_features tool automatically pulls domains, binding sites, and signal peptides, saving you hours of manual annotation review.
Validating protein activity used to be a nightmare. Now, calling get_proteomics or get_proteomics_ptm delivers actual mass-spectrometry evidence for modifications at the residue level.
See it in action
Assessing drug targets for cancer research
A geneticist finds a promising mutation in TP53. Instead of checking ClinVar, then running a separate search, they use get_variation and search_variation to immediately cross-reference the variant's clinical significance and population frequency across multiple sources.
Designing an antibody against a novel protein
A structural biologist wants to know which part of the target protein is best for binding. They first use get_protein_features to identify all potential domains, then run get_antigen to narrow down the optimal epitope region.
Integrating proteomics into a metabolomics pipeline
A researcher needs to validate protein expression levels. They retrieve the full protein record using get_protein and then call get_proteomics to see which peptides were actually detected via mass spectrometry in their samples.
Understanding species differences for a conserved pathway
A bioinformatician needs to compare human and mouse versions of the same enzyme. They use search_taxonomy first, then get_proteome on both IDs to quickly compare total protein counts and general architecture.
The honest tradeoffs
Searching by vague keywords
Trying to find variant data just by searching 'cancer mutation' without specifying the target gene or source.
Don't search generally. Instead, use get_variation or search_variation and specify the consequence type (e.g., missense) and cross-reference it with a known protein accession from get_protein.
Confusing general searches with specific data
Using search_proteins to find a list, then trying to use the result in a structural query like get_protein_features.
Always retrieve the full record first. Use get_protein(UniProt ID) to ensure you have the exact accession needed for detailed queries like getting sequence features.
Forgetting coordinate mapping
Getting a protein list and assuming it's already mapped to human chromosomes.
You must explicitly call get_coordinates. This function maps the protein ID to its exact genomic position on assemblies like GRCh38.
When It Fits, When It Doesn't
Use this MCP if your question involves synthesizing data from multiple, distinct biological domains: genetics (variants), structure (domains/sites), and experimental evidence (proteomics). For example, you need to know the sequence features of a protein AND its associated clinical variants. Don't use it if you just need general background reading or simple literature review; those are better handled by standard search engines. If your goal is purely taxonomic identification before any biology, start with search_taxonomy. But for deep-dive functional analysis—the kind that requires mapping coordinates to variant data and then checking binding sites—this MCP is necessary.
Questions you might have
How do I find all known variants for a protein? (using get_variation) +
You run get_variation with the UniProt accession. This tool pulls variant data from multiple sources like ClinVar and gnomAD, giving you aggregated results on clinical significance.
What is the difference between search_proteins and get_protein? +
Use search_proteins when you only know a keyword or an organism name. Use get_protein when you already have the precise UniProt accession ID for the specific protein record.
Can I find out where a gene is located on a chromosome? (using get_coordinates) +
Yes, use get_coordinates to map a protein. It returns the exact genomic location, including Ensembl IDs and the start/end positions on chromosomes like GRCh38.
How do I check for structural modifications? (using get_proteomics_ptm) +
Use get_proteomics_ptm. This tool delivers residue-level positions and evidence counts, showing exactly where post-translational modifications were detected in mass spectrometry data.
I need to find the correct species ID first. (using search_taxonomy) +
Start with search_taxonomy. This tool accepts common names or IDs and returns the precise NCBI taxon ID you need before querying anything else for that organism's proteome.
When using the tool `search_features_by_type`, what types of protein characteristics can I filter for? +
It returns a predefined list of feature categories like DOMAIN, BINDING, ACTIVE_SITE, and SIGNAL. You can narrow your search to specific regions—for instance, looking only for TRANSMEM or CARBOHYD features on a given protein.
What is the benefit of using `get_genecentric` over just fetching a full protein entry? +
It provides a gene-centric view by showing both the canonical protein count and related protein counts for that specific gene. This context helps you understand how important or prevalent that particular gene is within its entire proteome.
If I want an overview of all available data for an organism, should I use `search_proteomes`? +
Yes, using search_proteomes gives you essential summary metrics. It provides the proteome IDs, protein counts, and gene counts, letting you quickly gauge the scope of the reference data available for a given species.
Do I need an API key to use this server? +
No. The EMBL-EBI Proteins API is completely public and requires no authentication. Simply subscribe to this server and enter any placeholder value in the API key field to start querying protein data immediately.
What kind of variant data is available? +
The server aggregates genetic variants from multiple authoritative sources: UniProtKB curated variants, ClinVar clinical significance data, gnomAD population frequencies, 1000 Genomes Project, COSMIC somatic mutations, TOPMed whole-genome sequencing, ExAC exome data, and TCGA cancer variants. Each variant includes consequence type, clinical significance, and source cross-references.
Can I map protein positions to genome coordinates? +
Yes. The get_coordinates tool maps any UniProt protein to reference genome coordinates on GRCh38 and GRCh37 assemblies. It returns Ensembl gene, transcript, and translation identifiers along with chromosome, start/end positions, and strand orientation. This bridges the gap between protein-level annotations and genomic-level analyses.
We've already built the connector for EBI Proteins API. 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.