Metadata-Version: 2.4
Name: panhumanpy
Version: 1.0.0
Summary: A Python package for automated cell type annotation in scRNA-seq using Azimuth Neural Network.
License: MIT
Requires-Python: >=3.9
Requires-Dist: anndata
Requires-Dist: jupyter>=1.0.0
Requires-Dist: numpy
Requires-Dist: pandas
Requires-Dist: plotly
Requires-Dist: scikit-learn==1.6.0
Requires-Dist: scipy
Requires-Dist: tensorflow==2.17
Requires-Dist: umap-learn
Requires-Dist: xarray
Provides-Extra: gpu
Requires-Dist: tensorflow[and-cuda]==2.17; extra == 'gpu'
Description-Content-Type: text/markdown

# panhumanpy

**Current version: 1.0.0 (Orion)**

A package for unified and scalable organism-wide cell annotation in single-cell and spatial transcriptomics.

Uses the Pan-human Azimuth neural network.

## Prerequisites

- `python` >=3.9
- `pip`
- `git`

## Installation

To install the base version of the package (with CPU support only), run:

```bash
pip install panhumanpy
```

or to install from GitHub, run:

```bash
pip install git+https://github.com/satijalab/panhumanpy.git
```

If you require GPU acceleration for enhanced performance on compatible hardware, install the package with GPU dependencies:

```bash
pip install panhumanpy[gpu]
```

or from GitHub:


```bash
pip install git+https://github.com/satijalab/panhumanpy.git#egg=panhumanpy[gpu]
```

## Model Versions

panhumanpy uses versioned models corresponding to major package releases. The package defaults to model 'v{i}' where i is the major package version. For panhumanpy 1.0.0 (Orion), the default model version is 'v1'. For most users, the default version is recommended. The user can also choose to use a different model version as outlined in the tutorial mentioned below. 

Currently available model versions: 'v0', 'v1'

## Model Weights

Model weights are hosted on Zenodo and downloaded automatically on first use, cached in `~/.cache/panhumanpy/`. No manual download is required.

| Field | Detail |
|---|---|
| DOI | https://doi.org/10.5281/zenodo.20401417 |
| Models | v0, v1 |
| License | CC BY 4.0 |

## Cell Ontology Mapping

panhumanpy includes a built-in crosswalk that maps Pan-human Azimuth cell type annotations to [Cell Ontology](https://obofoundry.org/ontology/cl.html) (CL) terms. This mapping is versioned alongside the model and can be applied to annotation outputs via the `map_to_cell_ontology` function in `ANNotate_tools` or the `map_to_cell_ontology` method on `AzimuthNN` and `AzimuthNN_base`.

**Crosswalk provenance:**

| Field | Detail |
|---|---|
| Title | Crosswalk of Pan-human Azimuth Types annotated cells to Cell Ontology |
| Author | Aleix Puig-Barbe |
| Author ORCID | 0000-0001-6677-8489 |
| Reviewers | Bruce Herr II, Katy Borner, Jie Zheng |
| Reviewer ORCIDs | 0000-0002-6703-7647, 0000-0002-3321-6137, 0000-0002-2999-0103 |
| Data DOI | https://doi.org/10.48539/HBM727.TLKL.237 |
| Date | December 15, 2025 |
| Version | v1.1 |

## Tutorial

For an introductory tutorial, please check out this [notebook](https://github.com/satijalab/panhumanpy/blob/main/tutorial_panhumanpy.ipynb).

For more information on Pan-human Azimuth please visit our [website](https://satijalab.org/pan_human_azimuth/).
