Installation Guide¶

This section explains how to install and use the yomix tool.

Install via pip¶

In a Python virtual environment, do the following:

Install yomix using pip:
```
pip install yomix
```
Try the tool with the example dataset:
```
yomix --example
```

To use it on your own files:¶

Ensure your input file is an AnnData object saved in .h5ad format (see AnnData - Annotated Data <https://anndata.readthedocs.io/en/latest/>) with at least one .obsm field of dimension 2 or more.
Run the following command:
```
yomix yourfile.h5ad
```

Where yourfile.h5ad is your own dataset in .h5ad format.

Improve reactiveness with subsampling¶

When there are many samples in the dataset, you can use the –subsampling option to improve reactiveness. This option will randomly subsample the dataset to a maximum number of N samples. For example:

yomix --subsampling 5000 yourfile.h5ad

This will subsample the dataset to a maximum of 5000 samples.

Install from Source¶

If you prefer to install yomix from source, follow these steps:

Clone the repository:
```
git clone https://github.com/perrin-isir/yomix.git
```
We recommend using micromamba to create a Python environment, but you can use any Python package manager instead.
Navigate to the yomix directory:
```
cd yomix
```

Create the environment using micromamba:

micromamba create --name yomixenv --file environment.yaml

Activate the environment:
```
micromamba activate yomixenv
```
Install the package in editable mode:
```
pip install -e .
```
Try the tool with an example file:
```
yomix yomix/example/pbmc.h5ad
```

The input file must be an AnnData object saved in .h5ad format with at least one .obsm field of dimension 2 or more.

Notes¶

The input file must be an .h5ad file containing an AnnData object with at least one .obsm field of dimension 2 or more (e.g., PCA, UMAP embeddings).