lobster-bio-use
by cewinharhar
Git & GitHubBrief
Analyze biological data using Lobster AI — single-cell RNA-seq, bulk RNA-seq, literature mining, dataset discovery.
Description
What This Skill Does
Lobster AI is a multi-agent bioinformatics platform for single-cell and bulk RNA-seq analysis, literature mining, and visualization. Natural language queries are routed automatically to specialist agents handling QC, clustering, differential expression, annotation, and plotting. Data can be loaded from local files or fetched directly from GEO and SRA accessions.
Natural language routing across specialist agents removes the need to write separate analysis scripts for each bioinformatics step.
When to Use It
- Downloading a GEO dataset and running quality control in one step
- Clustering scRNA-seq cells and finding marker genes per cluster
- Searching PubMed for papers on a gene or disease
- Running differential expression between two cell populations
- Generating UMAP or volcano plots from processed single-cell data
Example Workflow
User asks: Download GSE109564, run QC, cluster the cells, and show a UMAP colored by cell type
- 1
Data Expert downloads GSE109564 from GEO
- 2
Transcriptomics Expert runs quality control and reports metrics
- 3
Transcriptomics Expert filters, normalizes, and clusters the cells
- 4
Annotation Expert identifies cell types and extracts marker genes
- 5
Visualization Expert generates a UMAP colored by cell type, saved as PNG and interactive HTML
Processed H5AD file, marker gene CSV, and UMAP visualization saved to the workspace
Requirements
- Lobster AI installed and initialized — run the platform installer from lobsterbio.com, then `lobster init` to configure API keys
Installation
clawhub install cewinharhar/lobster-bio-useRun the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.