Evo2 MCP Server
# evo2-mcp
[](https://biocontext.ai/registry/not-a-feature/evo2-mcp)
[![Tests][badge-tests]][tests]
[![Documentation][badge-docs]][documentation]
[badge-tests]: https://img.shields.io/github/actions/workflow/status/not-a-feature/evo2-mcp/test.yaml?branch=main
[badge-docs]: https://img.shields.io/readthedocs/evo2-mcp
The evo2-mcp server exposes [Evo 2](https://github.com/ArcInstitute/evo2) as a Model Context Protocol (MCP) server, providing tools for genomic sequence analysis. Any MCP-compatible client can use these tools to score, embed, and generate DNA sequences.
## Features
- **Sequence Scoring**: Compute log probabilities for DNA sequences
- **Sequence Embedding**: Extract learned representations from intermediate model layers
- **Sequence Generation**: Generate novel DNA sequences with controlled sampling
- **Variant Effect Prediction**: Score SNP mutations for variant prioritization
- **Multiple Model Checkpoints**: Support for 7B, 40B, and 1B parameter models
## Getting Started
**Prerequisites**: Python 3.12
1. **Install Evo2 dependencies**: See [Installation Guide][installation] for details.
```bash
conda install -c nvidia cuda-nvcc cuda-cudart-dev
conda install -c conda-forge transformer-engine-torch=2.3.0
pip install flash-attn==2.8.0.post2 --no-build-isolation
pip install evo2
```
2. **Install evo2-mcp**:
```bash
pip install evo2-mcp
```
3. **Activate MCP Server**:
Add the following to your `mcp.json` configuration:
```json
{
"mcpServers": {
"evo2-mcp": {
"command": "python",
"args": ["-m", "evo2_mcp.main"]
}
}
}
```
For detailed installation instructions, see the [Installation Guide][installation].
## Usage
Once installed, the server can be accessed by any MCP-compatible client. For available tools and usage examples, see the [Tools Documentation][tools].
### Available Tools
- `score_sequence` - Evaluate DNA sequence likelihood
- `embed_sequence` - Extract feature representations
- `generate_sequence` - Generate novel DNA sequences
- `score_snp` - Predict variant effects
- `get_embedding_layers` - List available embedding layers
- `list_available_checkpoints` - Show supported model checkpoints
See the [Tools Documentation][tools] for detailed API reference and examples.
## Documentation
- **[Installation Guide][installation]** - Detailed installation instructions
- **[Tools Reference][tools]** - Complete API documentation and usage examples
- **[Development Guide][development]** - Contributing and testing information
- **[Changelog][changelog]** - Version history and updates
You can also find this project on [BioContextAI](https://biocontext.ai/registry/not-a-feature/evo2-mcp), the community hub for biomedical MCP servers.
## Citation
If you use evo2-mcp in your research, please cite:
```bibtex
@software{evo2_mcp,
author = {Kreuer, Jules},
title = {evo2-mcp: MCP server for Evo 2 genomic sequence operations},
year = {2025},
url = {https://github.com/not-a-feature/evo2-mcp},
version = {0.2.3}
}
```
For the underlying Evo 2 model, please also cite the original Evo 2 publication.
## License and Attribution
The banner image in this repository is a modified version of the original [Evo 2 banner](https://github.com/ArcInstitute/evo2/blob/main/evo2.jpg) from the [Evo 2 project](https://github.com/ArcInstitute/evo2), which is released under the [Apache 2.0 License](https://www.apache.org/licenses/LICENSE-2.0). It was modified using Google Gemini "Nanobana" and GIMP.
[installation]: https://evo2-mcp.readthedocs.io/en/latest/installation.html
[tools]: https://evo2-mcp.readthedocs.io/en/latest/tools.html
[development]: https://evo2-mcp.readthedocs.io/en/latest/development.html
[issue tracker]: https://github.com/not-a-feature/evo2-mcp/issues
[tests]: https://github.com/not-a-feature/evo2-mcp/actions/workflows/test.yaml
[documentation]: https://evo2-mcp.readthedocs.io
[changelog]: https://evo2-mcp.readthedocs.io/en/latest/changelog.html
[pypi]: https://pypi.org/project/evo2-mcp
TDQS
Scored across 6 tools
Each tool targets a distinct operation: checkpoint listing, layer retrieval, sequence scoring, embedding extraction, sequence generation, and SNP scoring. Even score_sequence and score_snp are clearly differentiated by purpose and input requirements. No overlapping boundaries exist that would cause misselection.
All tools follow a verb_noun pattern in snake_case with clear, domain-specific nouns (list_available_checkpoints, get_embedding_layers, score_sequence, embed_sequence, generate_sequence, score_snp). The verbs accurately reflect the action, and the minor variation between 'list' and 'get' for retrieval is acceptable.
Six tools is a well-scoped set for an Evo 2 model server, covering discovery, feature extraction, scoring, generation, and variant analysis. Each tool earns its place without redundancy or unnecessary bulk.
The tool surface covers the core capabilities of the Evo 2 model: listing checkpoints, getting embedding layers, scoring sequences, extracting embeddings, generating sequences, and scoring SNP effects. No obvious gaps exist for standard use cases, and the inclusion of supporting discovery tools completes the workflow.