scRNA Agent

Developing a single cell RNAseq Multi-Agent

That can perform the following process:

1. scRNAseq Data Curation and Download
2. Preprocessing
3. Differential
4. Plots (eg. Volcano Plot, MA plot)
5. Visualization

scRNAseq Data Curation and Download

An LLM-assisted agent to search, evaluate, and download relevant single-cell RNA-seq datasets from GEO based on a biological query.

Features

• Query GEO using natural language or structured search terms
• Identify single-cell RNA-seq datasets
• Prefer processed data formats (MTX, TSV, CSV)
• Automatically download supplementary files
• Provide dataset-level biological context summaries

Example Usage

query = (
    "single cell[All Fields] AND RNA-seq[All Fields] AND immune[All Fields]"
)

final_output = run_agent(query)
print("\nFINAL OUTPUT:\n", final_output)

Processing

The agent performs the following steps:

1. Search GEO
2. Identify relevant datasets
3. Download processed data
4. Summarize biological relevance

Output

Downloaded supplementary files are stored locally: