Structures · Database
ESM Atlas
A database of 59.7 million metagenomic protein structure predictions covering proteins absent from AlphaFold DB.
What ESM Atlas is
The ESM Metagenomic Atlas is a database developed by Meta AI containing 59.7 million predicted protein structures.
The repository addresses the lack of structural information for uncharacterised metagenomic sequences, cataloguing the 'dark' protein universe that is absent from AlphaFold DB.
Key facts
- Records
- 59.7M structures
- Category
- structure
- Data type
- Metagenomic structure predictions
What you can use it for
- Metagenomic structural analysis
- Examining predicted three-dimensional structures for proteins identified from metagenomic samples.
- Exploring the dark protein universe
- Investigating structural folds for uncharacterised proteins that are missing from AlphaFold DB.
- Complementing reference structural databases
- Finding predicted models for sequences that do not appear in conventional protein structure databases.
Query it here
The research agent queries ESM Atlas directly — no account, no API key — and every answer links back to the record it came from. Ask in plain language and it writes the query.
Strengths and limits
Good at
- Supplies 59.7 million predicted protein structures from metagenomic sources.
- Provides structural coverage for parts of the dark protein universe omitted by AlphaFold DB.
Questions
Can I search ESM Atlas without writing code?
Yes. The research agent queries ESM Atlas directly, with no account or API key, and cites the record behind every answer. Ask in plain language — for a gene, a protein, a compound or a study — and it builds the query for you.
Can I reuse data from ESM Atlas?
MIT weights and code; the hosted endpoint adds its own fair-use terms. MIT covers the model. It does not cover unlimited use of someone else's free server. Terms: https://esmatlas.com/about
Who created the ESM Atlas?
The ESM Metagenomic Atlas was developed by Meta AI.
How many structures are catalogued in the database?
The resource contains 59.7 million predicted structures.
How does the database relate to AlphaFold DB?
It provides metagenomic structure predictions representing the dark protein universe that are absent from AlphaFold DB.
Written from ESM Atlas's own website and reference material. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.