Documentation¶

torch-adata is a framework for bridging data held in AnnData - the most popular single-cell python-based data structure and companion to scanpy - to the PyTorch Dataset class. torch-adata is meant to be structured but flexible within the rules of both of these data structures while also being easy to use.

Deep learning enthusiasts: get started faster with single-cell data.* Biologists: get started faster with ML
Store and access single-cell data efficiently and share between models, reproducibly (echoing the goals of the LightningDataModule.

Install torch-adata¶

To install the PYPI distribution:

pip install torch-adata

to install the development version:

git clone https://github.com/mvinyard/torch-adata.git; cd torch-adata
pip install -e .

Note

This project is under active development.

About the developer (Michael Vinyard)¶

I’m currently working on my PhD at Harvard University where I am developing methods for the analysis of single-cell genomics data. I also work on methods for CRISPR-Cas9-based genome-editing tools (like base-editing) to study cancer progression. My PhD work is funded by an NIH F31 award through the NCI and focuses on applying innovations in deep learning to analogous problems in the analysis of single-cell data.

If you find this package useful for your own work, please consider reaching out and connecting through GitHub, LinkedIn, or Twitter.