Metadata-Version: 2.5
Name: altar-alphagenome
Version: 0.8.0
Summary: AlphaGenome hosted-API model binding for inline Altar variant scoring
Project-URL: Documentation, https://kundajelab.github.io/altar/
Author: Riya Sinha
License-Expression: MIT
License-File: LICENSE
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
Classifier: Typing :: Typed
Requires-Python: >=3.12
Requires-Dist: alphagenome==0.9.0
Requires-Dist: altar<0.9,>=0.8.0
Provides-Extra: test
Requires-Dist: pytest-asyncio>=0.24; extra == 'test'
Requires-Dist: pytest>=8; extra == 'test'
Description-Content-Type: text/markdown

# altar-alphagenome

Adds Google DeepMind's hosted [AlphaGenome](https://github.com/google-deepmind/alphagenome) API to Altar: predicted
effects of variants on the human (hg38) or mouse (mm10) tracks and tissues you select, one API call per variant.

Validation: Not validated (hosted API; no validation record). The default prioritization rule is a screening rule,
not a calibrated threshold.

```bash
pip install altar-alphagenome
```

No resources to fetch. Export an AlphaGenome API key before the run:

```bash
export ALPHAGENOME_API_KEY=...
```

A run spec section; the spec names the variable, never the key:

```toml
[models.alphagenome]
type = "ALPHAGENOME"

[models.alphagenome.configuration]
credential = { provider = "env", name = "ALPHAGENOME_API_KEY" }
ontology_terms = ["UBERON:0002107"]   # liver; narrow the tracks to your tissues
scorers = ["ATAC"]
```

`altar models alphagenome` lists the columns, the `track_scores` detail table and the configuration fields.

Documentation: https://kundajelab.github.io/altar/integrations/alphagenome/

## Cite and terms

- AlphaGenome: Avsec Ž, et al. Nature 649, 1206–1218 (2026),
  [doi:10.1038/s41586-025-10014-0](https://doi.org/10.1038/s41586-025-10014-0).
- The API is subject to the [AlphaGenome Terms of Service](https://alphagenome.google/terms): free for
  non-commercial use, outputs not for training other models or for clinical decisions. Quota and availability
  apply. The `alphagenome` SDK is licensed Apache 2.0.

## Licence

MIT. See [LICENSE](https://github.com/kundajelab/altar/blob/main/bindings/alphagenome/LICENSE).
