Overview

BayesTME is a holistic Bayesian approach to the end-to-end modeling of ST data.

Please see our preprint

The figure below overviews the full BayesTME workflow and outputs:

BayesTME Pipeline

It is expected that the end user will run these 4 steps in sequence on their sample, but they can also be composed and mixed and matched. For example if bleeding correction was not relevant or necessary, it could be skipped and the pipeline started at phenotype selection. BayesTME aims to keep these four components modular to allow flexible workflows when necessary.

BayesTME provides a Command Line Interface that can be used to run each of the four steps, and subsequent creation of figures, but there are also main methods for each step that can be called to run the steps in your own python script, IPython notebook, or python based workflow manager.

If you use this package, please cite:

BayesTME: A unified statistical framework for spatial transcriptomics
H. Zhang, M. V. Hunter, J. Chou, J. F. Quinn, M. Zhou, R. White, and W. Tansey
bioRxiv 2022.07.08.499377.