A Python SDK developed by Google DeepMind for interacting with and visualizing genomic models. Currently at version 0.6.1, it supports interval scoring, in-silico mutagenesis, variant normalization, and sequence logo plotting. Requires Python >=3.10 and has a rapid release cadence (minor versions every few weeks).
pip install alphagenomeNo compatibility data collected yet for this library.
Verified import paths — ran on the pinned version, not inferred.
Initialize a model, predict tracks for a genomic interval.
Use intervals >= 10kb or set a larger width. Remove any hardcoded 2kb assumptions.
After upgrading to v0.6.0, verify that filtering by strand works as expected for both tracks and junctions.
Set the environment variable before using AlphaGenome: os.environ['ALPHAGENOME_API_KEY'] = 'your_key'