Loading connector details…
Loading connector details…
Choose a unique username to continue using AgentHotspot
by longevity-genie • Healthcare & Bioinformatics
An MCP server that exposes BioThings data (genes, variants, chemicals, taxons) to AI agents via the Model Context Protocol for standardized, typed biomedical queries and downloads.
Look up gene annotations and identifiers programmatically (using GeneTools) to enrich biomedical conversation or pipelines.
Retrieve and interpret variant annotations (using VariantTools) for clinical research or variant interpretation workflows.
Download sequences or formatted data files (FASTA, GenBank, JSON) and save them locally for downstream analysis or reporting.
This server implements the Model Context Protocol (MCP) for BioThings, providing structured programmatic access to mygene.info, myvariant.info, mychem.info and taxon data. It offers typed client interfaces (GeneTools, VariantTools, ChemTools, TaxonTools) plus DownloadTools for fetching and saving sequence and annotation files locally. The server supports multiple runtimes (stdio, HTTP, SSE) and integrates with MCP-compatible AI clients and inspectors to enable natural-language-driven, type-safe biomedical data retrieval and analysis.
Search genes via Lucene query, returning gene details and query metadata. **IMPORTANT:** This endpoint requires structured queries using specific field names. Simple natural language queries like "CDK2 gene" or "human kinase" will **NOT** work. You **MUST** specify the field you are querying, e.g., `symbol:CDK2`, `name:"cyclin-dependent kinase 2"`, `taxid:9606`. Use this tool when you need to *search* for genes based on criteria, not when you already know the specific gene ID. If you know the exact Entrez or Ensembl ID, use the `get_gene` tool instead for faster retrieval. **Supported Query Features (based on Lucene syntax):** 1. Simple Term Queries: `q=cdk2` (Searches across default fields) 2. Fielded Queries: `q=symbol:CDK2`, `q=name:"cyclin-dependent kinase 2"` 3. Range Queries: `q=taxid:[9606 TO 10090]` 4. Boolean Queries: `q=symbol:CDK2 AND taxid:9606` 5. Wildcard Queries: `q=symbol:CDK*` Returns gene details including symbol, name, taxid, and entrezgene.
Batch query genes by multiple terms, returning a list of gene details. Perform multiple gene searches in a single request using a comma-separated list of query terms. Unlike `query_genes`, the `query_list` parameter takes multiple **terms** (like gene IDs, symbols, names) rather than full query strings. The `scopes` parameter defines which fields these terms should be searched against. **Endpoint Usage:** - Query multiple symbols: `query_list=CDK2,BRCA1` with `scopes=symbol` - Query multiple Entrez IDs: `query_list=1017,672` with `scopes=entrezgene` - Query mixed IDs/symbols: `query_list=CDK2,672` with `scopes=symbol,entrezgene`
Fetch a specific gene by Entrez or Ensembl ID. Retrieves detailed information for a **single, specific gene** using its exact known identifier. **This is the preferred tool over `query_genes` for fetching a specific gene when you already know its standard ID (Entrez or Ensembl) and don't need complex search filters.** **Supported Identifiers:** - Entrez Gene ID: e.g., `1017` - Ensembl Gene ID: e.g., `ENSG00000123374`
Fetch multiple genes by a comma-separated list of Entrez or Ensembl IDs. Retrieves detailed information for **multiple specific genes** in a single request using their exact known identifiers. **This is the preferred tool over `query_many_genes` for fetching multiple specific genes when you already know their standard IDs (Entrez, Ensembl) and don't need complex search filters.** **Input Format:** Accepts a comma-separated list of gene IDs (Entrez or Ensembl). **Example:** `gene_ids=1017,1018` or `gene_ids=ENSG00000123374,ENSG00000134057`
Retrieve MyGene.info database metadata including stats and fields. Retrieve metadata about the underlying MyGene.info gene annotation database, **NOT** information about specific genes. **Use this tool ONLY to understand the database itself** (e.g., to discover available fields, check data versions, or get overall statistics). It **CANNOT** be used to find or retrieve data for any particular gene. **Returned Information:** - `stats`: Database statistics (e.g., total number of genes) - `fields`: Available gene annotation fields and their data types - `index`: Information about the backend data index - `version`: Data version information
Scores are informational only and provided “as is” without warranty. AgentHotspot assumes no liability for actions taken based on these ratings.