Claw4Science

Getting Started with BioClaw: Run Bioinformatics from WhatsApp

Mar 19, 2026

What is BioClaw?

BioClaw brings computational biology directly into your messaging apps. Instead of SSH-ing into a server, opening a Jupyter notebook, or memorizing CLI flags, you simply type a natural language request in a WhatsApp group chat — and BioClaw handles the rest.

Built on the NanoClaw container architecture with bioinformatics tools from the STELLA project (Princeton/Stanford/UCLA), BioClaw wraps a full computational environment inside Docker and exposes it through the Claude Agent SDK.

What Can You Do With It?

BioClaw comes with a rich set of preinstalled tools:

Command-line tools: BLAST+, SAMtools, BEDTools, BWA, minimap2, FastQC, MultiQC, seqtk, fastp, bcftools, PyMOL, salmon, kallisto, and more.

Python libraries: BioPython, pandas, NumPy, SciPy, matplotlib, seaborn, scikit-learn, RDKit, PyDESeq2, scanpy, pysam.

All accessible through natural language — no command memorization needed.

Quick Start

Prerequisites

  • macOS or Linux (Windows users: see WSL2 guide)
  • Node.js 20+
  • Docker Desktop (running)
  • An Anthropic API key or OpenRouter API key

Installation

git clone https://github.com/Runchuan-BU/BioClaw.git
cd BioClaw
npm install
npm start

On first launch, a QR code will appear in the terminal. Scan it with WhatsApp to connect. Authentication persists in store/auth/.

Model Configuration

Create a .env file in the project root. Choose one provider:

Option A: Anthropic (default)

ANTHROPIC_API_KEY=your_anthropic_key

Option B: OpenRouter (supports Gemini, DeepSeek, Claude, GPT)

MODEL_PROVIDER=openrouter
OPENROUTER_API_KEY=sk-or-v1-your-key
OPENROUTER_BASE_URL=https://openrouter.ai/api/v1
OPENROUTER_MODEL=deepseek/deepseek-chat-v3.1

Popular model IDs: deepseek/deepseek-chat-v3.1, google/gemini-2.5-flash, anthropic/claude-3.5-sonnet. Make sure the model supports tool calling.

Channels

BioClaw supports multiple messaging platforms:

  • WhatsApp — default, QR code auth
  • WeCom (Enterprise WeChat) — API bot mode with long polling
  • Discord — standard bot with message content intent
  • Local Web UI — set ENABLE_WHATSAPP=false and ENABLE_LOCAL_WEB=true, then open http://127.0.0.1:3210

Example Tasks

Here are 6 real demo tasks you can try once BioClaw is running. Just send these messages in your connected group chat:

1. Workspace Triage

@Bioclaw In /workspace/group/4, list the files and recommend the best next analysis steps. Keep it short (max 8 bullets) and ask me 1 question to proceed.

Workspace triage demo

2. FastQC Quality Check

@Bioclaw Make a quick QC summary from /workspace/group/reads_R1.fastq.gz and /workspace/group/reads_R2.fastq.gz (FastQC). Send me the key findings in bullets and also send the FastQC report image(s) to WhatsApp.

FastQC demo

@Bioclaw BLAST this sequence against nr (protein). Tell me the top 5 hits with species, % identity, e-value, and a one-line interpretation:
>query
MSTNPKPQRKTKRNTNRRPQDVKFPGG...

BLAST demo

4. Volcano Plot

@Bioclaw Create a simple volcano plot from /workspace/group/counts.csv (assume columns: gene, log2FC, pvalue). Use readable labels, export a PNG, and send the plot image to WhatsApp with a 2-sentence takeaway.

Volcano plot demo

5. Protein Structure Rendering

@Bioclaw Render the structure of PDB 1UBQ in rainbow colors, generate a high-resolution image, and send it to WhatsApp. Then give me 3 bullets describing what I'm looking at.

Protein structure demo

@Bioclaw Search PubMed for recent papers about ADHD published in high-impact journals and summarize the top three. For each paper: citation, why it matters, and 1 limitation.

PubMed search demo

Architecture

Chat Platform → Node.js Orchestrator → SQLite State → Docker Container → Agent + Bio Tools

BioClaw uses a containerized approach: each session runs inside an isolated Docker container with the full bioinformatics toolbox preinstalled. The Node.js orchestrator manages message routing, session state (SQLite), and container lifecycle.

Citation

If you use BioClaw in your research, please cite the STELLA paper:

@article{jin2025stella,
  title={STELLA: Towards a Biomedical World Model with Self-Evolving Multimodal Agents},
  author={Jin, Ruofan and Xu, Mingyang and ...},
  journal={bioRxiv},
  year={2025},
  doi={10.1101/2025.07.01.662467}
}