AI · Bioinformatics · Multi-Platform

BioClaw — the research
assistant biologists deserve.

BLAST searches, protein structure rendering, sequencing QC, literature mining — through natural language in any chat.

BioClaw · agent session
$ @BioClaw BLAST this sequence against nr
 
▶ Running BLAST+ ncbi/nr
  query_length=248aa · e-value=1e-5
 
✓ Top hits returned
  1. TP53_HUMAN  Score=520  E=0.0
  2. TP53_MOUSE  Score=498  E=0.0
  3. TP63_HUMAN  Score=312  E=2e-89
 
$
PyMOL render complete
FastQC report ready

Multi-Platform

Works wherever your team chats

One agent, every surface — real results delivered back to the conversation.

Channel config → docs/CHANNELS.md · Local web includes lab trace (SSE timeline) → docs/DASHBOARD.md


Capabilities

What BioClaw can do

Send a natural language message — BioClaw handles the rest, returning results directly to your chat.

🧬
01

Workspace Triage

Analyze files and recommend the best next analysis steps.

🔬
02

FastQC Quality Control

Run FastQC on FASTQ files and deliver the QC report.

🔍
03

BLAST Sequence Search

BLAST a protein sequence against NCBI nr database.

📊
04

Volcano Plot Generation

Create differential expression plots from a CSV file.

🧪
05

Protein Structure Rendering

Fetch a PDB structure, render with PyMOL, send image.

📚
06

PubMed Literature Search

Search PubMed and provide structured paper summaries.

⚗️
07

Hydrogen Bond Analysis

Visualize H-bonds between ligand and protein.

🎯
08

Binding Site Visualization

Show residues within 5Å of a ligand as an image.


Toolkit & skills

A full bioinformatics toolkit

Command-line tools and Python libraries pre-installed in Docker, plus 25 built-in skill guides for analyses, omics workflows, and public-database lookups — all reachable in natural language.

Agent skills

25+ runtime skills

Actively growing
🚀
More skills in development
New analysis workflows, database integrations, and visualization tools are continuously being added. Check back or star the repo to stay updated.

Architecture

How it works

Built on NanoClaw. Each group runs in an isolated Docker container with a full bioinformatics environment. Powered by the STELLA framework.

Chat ──► Node.js Orchestrator ──► SQLite ──► Docker Container
Claude Agent SDK
Bioinformatics Toolbox
BLAST+SAMtoolsBWAPyMOLFastQCminimap2scanpyPyDESeq2BioPythonpandasmatplotlibRDKit
Container Isolation — each group has its own Docker workspace
Filesystem IPC — results delivered back via filesystem
Channel Agnostic — Feishu · WeCom · Discord · Slack · WeChat · QQ · local web

Quick Start

Up and running in minutes

Requires Node.js 20+ and Docker Desktop. Supports Anthropic API and OpenRouter.

$git clone https://github.com/Runchuan-BU/BioClaw.git && cd BioClaw && npm install && npm start
Full setup guide → GitHub README ↗

Citation

Cite this work

BioClaw builds on the STELLA framework. If you use it in research, please cite the STELLA preprint.

BibTeX
@article{jin2025stella, title={STELLA: Towards a Biomedical World Model with Self-Evolving Multimodal Agents}, author={Jin, Ruofan and Xu, Mingyang and Meng, Fei and Wan, Guancheng and Cai, Qingran and Jiang, Yize and Han, Jin and Chen, Yuanyuan and Lu, Wanqing and Wang, Mengyang and Lan, Zhiqian and Jiang, Yuxuan and Liu, Junhong and Wang, Dongyao and Cong, Le and Zhang, Zaixi}, journal={bioRxiv}, year={2025}, doi={10.1101/2025.07.01.662467} }

Contact

Get in touch

Questions about BioClaw or this site — send a message below.

We'd love to hear from you
Research collaboration · Integration partnership · Investment inquiry · Bug report · Feature request · General questions — all welcome.
Apply as Co-creation Partner

Join our open co-creation program — help shape the future of BioClaw together.