Metadata-Version: 2.5
Name: altar-sei
Version: 0.5.0
Summary: Sei model binding for large-scale 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: altar<0.6,>=0.5.0
Provides-Extra: test
Requires-Dist: pytest-asyncio>=0.24; extra == 'test'
Requires-Dist: pytest>=8; extra == 'test'
Description-Content-Type: text/markdown

# altar-sei

Adds [Sei](https://github.com/FunctionLab/sei-framework) to Altar: predicted effects of human SNVs (hg19 or hg38)
on 40 sequence classes of regulatory activity, run in a container on an NVIDIA GPU or a CPU.

Validation: Upstream parity (CPU), against the official sei-framework.

```bash
pip install altar-sei
```

The binding has no resource command. Download `sei_model.tar.gz` from the official
[Zenodo record](https://doi.org/10.5281/zenodo.4906996) and extract it. The files are in its `model/` directory,
and the section below gives their official digests; `sha256sum` checks your copy.

A run spec section, with the official files' digests:

```toml
[models.sei]
type = "SEI"

[models.sei.configuration]
weights = { uri = "model/sei.pth", digest = "sha256:9d4771d5363e4955eec2a36567fedac986191f64c7874c85bfb9e203c7f60783" }
projection_vectors = { uri = "model/projvec_targets.npy", digest = "sha256:7080fd22e98b3bb47436801f94682d8b097a29f4d8c9cb154210e64e0e4190c8" }
histone_indices = { uri = "model/histone_inds.npy", digest = "sha256:7a7af2c642ae6833ce2cee799a702a907544c0926c56b5c8b1236cd6470f5443" }
sequence_class_names = { uri = "model/seqclass.names", digest = "sha256:5276a0a39f4d0364e22c5bb93d02e75c7c171f6e4b61fc47baa301fce9cc15b3" }
```

`altar models sei` lists the columns, the `sequence_class_effects` detail table, the opt-in
`chromatin_profile_effects` artifact and the configuration fields.

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

## Cite and terms

- Sei: Chen KM, et al. Nature Genetics 54, 940–949 (2022),
  [doi:10.1038/s41588-022-01102-2](https://doi.org/10.1038/s41588-022-01102-2).
- The official Sei code and model are licensed for academic and research use only. The Sei runtime image carries
  that licence. This binding contains no Sei code or weights.

## Licence

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