Multi-Agent Bioinformatics Analysis System powered by LangGraph
Project description
🦞 Lobster AI
Multi-agent bioinformatics engine. Analyze multi-omics data in natural language.
Documentation · PyPI · Omics-OS Cloud · Agent Template
What is Lobster AI?
Lobster AI is an open-source multi-agent platform for bioinformatics. Describe your analysis in natural language — Lobster routes it to specialized AI agents that handle single-cell RNA-seq, bulk RNA-seq, proteomics, genomics, literature mining, and machine learning. Results are reproducible (W3C-PROV), exportable as Jupyter notebooks, and backed by the same open-source tools researchers already trust (Scanpy, PyDESeq2, AnnData).
Built for bioinformaticians, computational biologists, and research teams who need intelligent workflows without writing code.
Quick Start
macOS / Linux (recommended):
curl -fsSL https://install.lobsterbio.com | bash
Windows (PowerShell):
irm https://install.lobsterbio.com/windows | iex
Manual (any platform):
uv tool install 'lobster-ai[full,anthropic]'
lobster init
lobster chat
Or with pip:
pip install 'lobster-ai[full]'
lobster init
lobster chat
Agent Packages
Lobster AI uses a modular architecture. Install exactly the agents you need.
| Package | Agents | Status |
|---|---|---|
| lobster-ai | Supervisor, Core infrastructure | Beta |
| lobster-transcriptomics | Transcriptomics, Annotation, DE Analysis | Beta |
| lobster-research | Research, Data Expert | Beta |
| lobster-visualization | Visualization Expert | Beta |
| lobster-metadata | Metadata Assistant | Beta |
| lobster-structural-viz | Protein Structure Visualization | Beta |
| lobster-genomics | Genomics Expert | Alpha |
| lobster-proteomics | Proteomics Expert | Alpha |
| lobster-ml | ML, Feature Selection, Survival Analysis | Alpha |
Add individual agents (uv tool installs):
uv tool install lobster-ai --with lobster-proteomics --with lobster-genomics
Or run lobster init --force to interactively select agents — it generates the command for you.
Install everything:
uv tool install 'lobster-ai[full,anthropic]' # uv tool (recommended)
pip install 'lobster-ai[full]' # pip
Upgrade:
uv tool upgrade lobster-ai # uv tool
pip install --upgrade lobster-ai # pip
Build your own agent: Use the lobster-agent-template to create custom analysis agents.
Coding Agent Skills
Teach your coding agents (Claude Code, Codex, Gemini CLI, OpenClaw) to use Lobster:
curl -fsSL https://skills.lobsterbio.com | bash
Installs two skills:
- lobster-use — End-user workflows (search PubMed, analyze cells, RNA-seq)
- lobster-dev — Developer guide (create agents, extend services, testing)
LLM Providers
Lobster supports 5 LLM providers. Choose based on your needs:
| Provider | Type | Setup | Use Case |
|---|---|---|---|
| Ollama | Local | ollama pull gpt-oss:20b |
Privacy, zero cost, offline |
| Anthropic | Cloud | API key | Fastest, best quality |
| AWS Bedrock | Cloud | AWS credentials | Enterprise, compliance |
| Google Gemini | Cloud | Google API key | Multimodal, long context |
| Azure AI | Cloud | Endpoint + credential | Enterprise Azure |
Configure via lobster init or set environment variables manually.
Features
Multi-Omics Analysis
- Single-cell RNA-seq (QC, clustering, cell type annotation, trajectory)
- Bulk RNA-seq (Kallisto/Salmon import, DE with PyDESeq2)
- Proteomics (DDA/DIA, missing values, normalization)
- Genomics (VCF/PLINK, GWAS, PCA, variant annotation)
- Multi-omics integration (MOFA, pathway enrichment)
Machine Learning
- Feature selection (stability selection, LASSO, variance filter)
- Survival analysis (Cox models, Kaplan-Meier, risk stratification)
- Cross-validation (stratified k-fold, nested CV)
- Interpretability (SHAP, feature importance)
Research & Metadata
- Literature discovery (PubMed, PMC, GEO)
- Dataset downloads (GEO, SRA, PRIDE)
- Metadata harmonization and validation
- Sample grouping and filtering
Reproducibility
- W3C-PROV provenance tracking
- Jupyter notebook export
- Parameter schemas for all analyses
Usage Examples
Interactive chat:
lobster chat
Single-turn queries:
lobster query "Search PubMed for CRISPR studies in 2024"
lobster query "Download GSE109564 and cluster cells"
Session continuity:
lobster query --session-id project1 "Search for Alzheimer's scRNA-seq datasets"
lobster query --session-id project1 "Download the top 3 results"
lobster query --session-id latest "Cluster the first dataset"
Export pipelines:
lobster chat
> /pipeline export
> /pipeline list
> /pipeline run analysis.ipynb geo_gse109564
Development
# Clone repository
git clone https://github.com/the-omics-os/lobster.git
cd lobster
# Install with dev dependencies (editable)
make dev-install
# Run tests
make test
# Format and lint
make format && make lint
# Activate environment
source .venv/bin/activate
# Test as end user (uv tool install)
uv tool install 'lobster-ai[full,anthropic]'
Environment variables (will be created during init):
ANTHROPIC_API_KEY— Anthropic DirectAWS_BEDROCK_ACCESS_KEY,AWS_BEDROCK_SECRET_ACCESS_KEY— AWS BedrockGOOGLE_API_KEY— Google GeminiOLLAMA_BASE_URL— Ollama server (default: http://localhost:11434)LOBSTER_LLM_PROVIDER— Explicit provider selectionLOBSTER_WORKSPACE— Workspace directory
Documentation
For detailed documentation, see README_FULL.md or visit docs.omics-os.com.
Key topics:
- Platform-specific installation (macOS, Linux, Windows)
- One-line installers (
install.lobsterbio.com) - uv tool install and upgrade workflows
- Docker deployment
- LLM provider configuration
- Premium features
- Troubleshooting
Ecosystem
| Project | Description | Link |
|---|---|---|
| Lobster AI | Open-source multi-agent bioinformatics engine | lobsterbio.com |
| Omics-OS Cloud | Managed cloud platform | app.omics-os.com |
| Documentation | Guides, API reference, tutorials | docs.omics-os.com |
| Agent Template | Create your own Lobster agent | GitHub |
Contributing
Contributions are welcome. See CONTRIBUTING.md for guidelines.
Ways to contribute:
- Report bugs and request features via GitHub Issues
- Submit pull requests for bug fixes or new features
- Improve documentation
- Create custom agent packages
License
Code: AGPL-3.0-or-later Documentation: CC BY 4.0
See LICENSE for details.
Built by Omics-OS to accelerate multi-omics research
Omics-OS · Lobster AI · Docs
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file lobster_ai-1.0.2.tar.gz.
File metadata
- Download URL: lobster_ai-1.0.2.tar.gz
- Upload date:
- Size: 1.1 MB
- Tags: Source
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.7
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
d44cc3cf402f4d0525c38f0e151b5d481dccc9004dbb24185c46cc8531a6270d
|
|
| MD5 |
49de629000c065e39b79f6c680b09362
|
|
| BLAKE2b-256 |
beb4fb0fdf68bbb780b72b0aab1a1be6e4777a01f9c4979f4c5ebae462a22cb3
|
Provenance
The following attestation bundles were made for lobster_ai-1.0.2.tar.gz:
Publisher:
publish-packages.yml on the-omics-os/lobster
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
lobster_ai-1.0.2.tar.gz -
Subject digest:
d44cc3cf402f4d0525c38f0e151b5d481dccc9004dbb24185c46cc8531a6270d - Sigstore transparency entry: 949229491
- Sigstore integration time:
-
Permalink:
the-omics-os/lobster@79411f90fe6d6ce8f7164c85d2e8e431321f4175 -
Branch / Tag:
refs/tags/v1.0.2 - Owner: https://github.com/the-omics-os
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish-packages.yml@79411f90fe6d6ce8f7164c85d2e8e431321f4175 -
Trigger Event:
push
-
Statement type:
File details
Details for the file lobster_ai-1.0.2-py3-none-any.whl.
File metadata
- Download URL: lobster_ai-1.0.2-py3-none-any.whl
- Upload date:
- Size: 1.3 MB
- Tags: Python 3
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.7
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
834d76aebfbc46f222fb5fcb227aedd4564ae0946e8501408c292fe03d3fbd3c
|
|
| MD5 |
ced2f8c9c17626d00314003cc77753f2
|
|
| BLAKE2b-256 |
e849163f8019c0d4d3443bd910fc88822847ccd4771d8401b1edbf603722ea7c
|
Provenance
The following attestation bundles were made for lobster_ai-1.0.2-py3-none-any.whl:
Publisher:
publish-packages.yml on the-omics-os/lobster
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
lobster_ai-1.0.2-py3-none-any.whl -
Subject digest:
834d76aebfbc46f222fb5fcb227aedd4564ae0946e8501408c292fe03d3fbd3c - Sigstore transparency entry: 949229547
- Sigstore integration time:
-
Permalink:
the-omics-os/lobster@79411f90fe6d6ce8f7164c85d2e8e431321f4175 -
Branch / Tag:
refs/tags/v1.0.2 - Owner: https://github.com/the-omics-os
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish-packages.yml@79411f90fe6d6ce8f7164c85d2e8e431321f4175 -
Trigger Event:
push
-
Statement type: