ensembl-database

Resolve gene symbols and Ensembl IDs, retrieve sequences, and predict variant effects via Ensembl REST APIs.

2.7k|283|Updated May 13, 2026
One-click install
npx skills add https://github.com/google-deepmind/science-skills --skill ensembl-database-google-deepmind
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ensembl-database
Source: https://github.com/google-deepmind/science-skills/tree/main/skills/ensembl_database
Command: npx skills add https://github.com/google-deepmind/science-skills --skill ensembl-database-google-deepmind

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill removes the manual work of converting gene/transcript/protein identifiers and looking up genomic context by letting you query Ensembl for ID mappings, sequences, transcript structures, and variant consequence predictions.

Core Features & Use Cases

  • Primary ID resolution and cross-references: Resolve gene symbols/aliases/RefSeq to ENSG IDs and map Ensembl IDs to external databases like UniProt, HGNC, and RefSeq.
  • Genomic and protein sequence retrieval: Fetch raw genomic DNA for coordinate windows and pull amino-acid FASTA for transcripts/proteins.
  • Transcript structures and variant effects: Retrieve exon/CDS/UTR layout for transcripts and run Ensembl VEP with open-license plugins (e.g., AlphaMissense, LOEUF, NMD), returning both stdout predictions and full JSON outputs.

Quick Start

Use the ensembl-database skill to predict consequences for a variant you provide as chr:pos:ref:alt, then return the full VEP transcript prediction list and save the complete JSON to a file.

Frequently Asked Questions about ensembl-database

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I predict variant effects using Ensembl VEP for a specific genomic coordinate?

To predict variant effects, you provide your input as chr:pos:ref:alt, rsID, or HGVS notation. The skill queries the Ensembl VEP REST API, returning a concise transcript prediction summary to stdout and saving the complete JSON output to a file.

Can I map gene symbols and RefSeq IDs to Ensembl stable IDs?

Yes, you can resolve ambiguous gene symbols, aliases, and RefSeq identifiers to Ensembl stable IDs. The skill cross-references these mappings to external databases like UniProt and HGNC via the Ensembl REST API.

Does this tool support both GRCh37 and GRCh38 genome assemblies?

Yes, the skill supports both GRCh37 and GRCh38 genome assemblies. It automatically directs queries to the correct Ensembl REST API base URL to ensure accurate genomic sequence retrieval and variant consequence prediction.

How do I retrieve exon, CDS, and UTR transcript structures from Ensembl?

You can retrieve transcript structures by querying the Ensembl REST API with a transcript ID. The skill models the exon, CDS, and UTR layout, providing a concise summary while saving the complete genomic context to a file.

What is the best way to fetch genomic DNA and protein FASTA sequences for a specific gene?

The best way to fetch sequences is to provide the gene or transcript ID to retrieve raw genomic DNA windows or amino-acid FASTA. The skill handles rate-limit friendly Ensembl REST API requests and outputs the FASTA results.

Can I use AlphaMissense and LOEUF plugins with Ensembl VEP predictions?

Yes, the skill runs Ensembl VEP with open-license plugins like AlphaMissense, LOEUF, and NMD. It computes variant consequence predictions and returns the full transcript prediction list with plugin annotations included.