Lobster Bioinformatics Agent

Lobster AI is a bioinformatics platform that combines specialized AI agents with open-source tools to analyze multi-omics data through natural language.

When to use this Skill

Use Lobster when the user asks to:

• Analyze single-cell RNA-seq data (QC, clustering, annotation, markers)
• Perform bulk RNA-seq analysis (differential expression, complex designs)
• Search scientific literature (PubMed, PMC, full-text retrieval)
• Discover datasets (GEO, SRA, ENA (free) and PRIDE, MASSive (cloud))
• Run quality control on biological data
• Generate bioinformatics visualizations (UMAP, volcano plots, heatmaps)
• Download and process biological datasets
• Work with H5AD, CSV, Excel, 10X formats
• Extract methods or metadata from papers

Requirements

Lobster must be installed and configured:

# Check if Lobster is installed
which lobster

# If not installed:
uv pip install lobster-ai
lobster init --help #to see non-interactive

Lobster requires an LLM provider (Ollama, Anthropic, or AWS Bedrock).

Pre-flight check (IMPORTANT)

Before running any analysis, always verify Lobster is ready:

lobster config-test --json

Returns structured JSON:

{
  "valid": true,
  "env_file": "/path/to/.env",
  "checks": {
    "llm_provider": {"status": "pass", "provider": "bedrock", "message": "Connected"},
    "ncbi_api": {"status": "pass", "has_key": true, "message": "Connected"},
    "workspace": {"status": "pass", "path": "/path/to/workspace", "message": "Writable"}
  }
}

This command validates:

• LLM provider - Ollama server running + models installed, or Anthropic/Bedrock API keys valid
• NCBI API - PubMed/GEO access (optional but recommended)