Single-cell analysis agent

## 🧬 Single-Cell Analysis Agent: AI-Powered scRNA-seq Assistant

### Inspiration

Single-cell RNA sequencing has revolutionized our understanding of cellular heterogeneity, but the computational barrier to entry remains high. Biologists and clinicians who generate the data often spend weeks learning Python, Scanpy, and command-line tools just to perform basic quality control and clustering. We built the Single-Cell Analysis Agent to democratize single-cell analysis — an AI agent that speaks your language, understands biology, and writes reproducible analysis code on the fly.

### What it does

The Single-Cell Analysis Agent is an intelligent assistant that transforms natural language prompts into complete single-cell analysis workflows. Instead of memorizing API calls, you simply describe what you want:

> "Filter out cells with >20% mitochondrial reads, normalize by total counts, find highly variable genes, and run PCA with 30 components."

The agent generates, executes, and explains the corresponding Scanpy code — all within an interactive chat interface. It handles:

- Quality control — automated filtering based on QC metrics
- Normalization & scaling — log-normalization, SCTransform, etc.
- Dimensionality reduction — PCA, UMAP, t-SNE
- Clustering — Leiden, Louvain with resolution tuning
- Differential expression — Wilcoxon, pseudobulk DESeq2
- Cell-type annotation — marker-based and reference-mapped
- Visualization — customizable UMAP plots, heatmaps, violin plots

Built With

• intent