Bering: Transfer Learning of Cell Segmentation and Annotation for Spatial Omics
Project description
Bering - Spatial Segmentation and Cell Annotation in Python
Bering is a deep learning algorithm for simultaneous molecular annotation and cell segmentation in single-cell spatial transcriptomics data.
It builds on top of torch_geometric
_ and scanpy
_, from which it inherits modularity and scalability.
It provides versatile models that leverages the spatial coordinates of the data, as well as pre-trained models across spatial technologies and tissues.
Visit our documentation
_ for installation, tutorials, examples and more.
Manuscript
The manuscript has been submittet for peer review. A preprint will be released soon.
Bering's key applications
- Identify background and real signals in noisy spatial transcriptomics data.
- Identify cell annotations for transcripts on single-cell spatial data.
- Efficiently cell segmentation with cell annotations.
- Build and fine-tune pre-trained model on new data using transfer learning.
Installation
Install Bering via PyPI by running::
pip install Bering
or via Conda as::
conda install -c conda-forge Bering
Contributing to Bering
We are happy about any contributions! Before you start, check out our contributing guide <CONTRIBUTING.rst>
_.
.. _Palla, Spitzer et al. (2022): https://doi.org/10.1038/s41592-021-01358-2 .. _scanpy: https://scanpy.readthedocs.io/en/stable/ .. _torch_geometric: https://pytorch-geometric.readthedocs.io/en/latest/ .. _documentation: https://celldrift.readthedocs.io/en/latest/index.html
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.