Documentation

Contents

Documentation#

scanvi-explainer is a package which helps to describe which features (genes) are required when predicting a cell type using SCANVI [Xu et al., 2021]. We utilize SHAP package [Lundberg and Lee, 2017] to estimate the importance of each gene.

This package follows best practices from scvi-tools [Gayoso et al., 2022].

Citation#

Please consider citing scANVI Explainer if you use in your research.

Deep Learning Based Models for Preimplantation Mouse and Human Development
Martin Proks, Nazmus Salehin, Joshua M. Brickman
bioRxiv 2024.02.16.580649; doi: 10.1101/2024.02.16.580649

@article{Proks2024.02.16.580649,
  author = {Proks, Martin and Salehin, Nazmus and Brickman, Joshua M.},
  title = {Deep Learning Based Models for Preimplantation Mouse and Human Development},
  elocation-id = {2024.02.16.580649},
  year = {2024},
  doi = {10.1101/2024.02.16.580649},
  publisher = {Cold Spring Harbor Laboratory},
  URL = {https://www.biorxiv.org/content/early/2024/02/16/2024.02.16.580649},
  eprint = {https://www.biorxiv.org/content/early/2024/02/16/2024.02.16.580649.full.pdf},
  journal = {bioRxiv}
}
Contents