Trellis tree-based analysis reveals stromal regulation of patient-derived organoid drug responses
This library runs code associated with the Trellis paper which is now published in Cell and accessible here.
A short how-to tutorial on Trellis is available here, with associated notebook hosted on Kaggle here.
Code to reproduce the figures in the paper can be found on the TapeLab's github repo. Note that an earlier version of MultiscaleEMD was used to run these experiments. Installing this version can be accomplished by downloading the source at this commit and installing with pip install -e . from the MultiscaleEMD directory.
In brief, Trellis is a method to compare single-cell dataset distributions under different treatment conditions while normalizing for multiple controls. Paired Trellis implements a Kantorovich-Rubenstein distance with tree ground distance allowing for a normalization step against a specified control per treatment.
Trellis is available in pypi. Install by running the following:
pip install MultiscaleEMDThis code was tested with python 3.7-3.10. Issues can be filed directly on the MultiscaleEMD project.
All mass cytometry files are available on Cytobank.
Compiled TOBis mass cytometry PDO-CAF dataframe is available on Mendely.
- @article{Ramos Zapatero2022.10.19.512668,
- author = {Maria {Ramos Zapatero} and Alexander Tong and James W. Opzoomer and Rhianna O'Sullivan and Ferran {Cardoso Rodriguez} and Jahangir Sufi and Petra Vlckova and Callum Nattress and Xiao Qin and Jeroen Claus and Daniel Hochhauser and Smita Krishnaswamy and Christopher J. Tape}, journal = {Cell}, number = {25}, pages = {5606-5619.e24}, title = {Trellis tree-based analysis reveals stromal regulation of patient-derived organoid drug responses}, volume = {186}, year = {2023}
}