Enhancer activity

Regression that predicts enhancer activity for a sequence, returning two continuous scores: developmental (dev) and housekeeping (hk). The default model (g0-deepstarr) is a G0 BERT-Base encoder trained on Drosophila enhancers from the DeepSTARR dataset.

Input

A DNA sequence passed as sequence (inline) or a server-side handle. The model uses a 249 bp window. Because the model is trained on Drosophila, use a fly sequence for meaningful results.

Output

Two per-window activity scores (dev, hk), returned in the {data, meta} envelope. Available output formats include bedgraph and json.

Try it

REST API: POST /v1/tasks/enhancer/predict — see the getting-started guide. MCP: "Fetch the Drosophila ftz gene and predict enhancer regions." See mcp.md.