Use chart selections in Python

Brush a chart to get selected regions or genes in Python. You can save those results or send new annotations back to the chart.

Start with an interactive workflow:

• Annotate genomic intervals: name regions, add them to a separate track, and export BED.

• Select genes for follow-up: brush a volcano plot, inspect the selected genes, and export a CSV.

How it works

Each notebook connects the chart to Python in a background task. Let the cell finish, then interact with the chart. The connection keeps listening while you work.

The annotation notebook stores records in annotations. The gene-selection notebook updates a pandas table called selected_genes:

# Run after brushing in the gene-selection notebook.
selected_genes[["ensgene", "log2fc", "padj"]]

These values stay in memory until you export them. Rerun an inspection cell to refresh its output; the tables beside the charts update automatically.

Note

The web demos run in your browser. To update Python values, use a running kernel in Jupyter or VS Code and wait for the notebook’s connection message.

If it does not connect

Restart the kernel and run all cells. Keep the background-task setup from the example: waiting for a browser reply directly in a cell can block that reply in VS Code.

The embed API is experimental. See advanced integration for supported operations, custom hosts, and connection details.