They use bayesian inference to model differential expression from a CoDA perspective in single-celled datasets. Not sure if you've come across it but thought that it could be useful since you guys are currently developing BIRDMAn. scCODA is built on tensorflow 2 and tensorflow probability.
https://www.nature.com/articles/s41467-021-27150-6
https://github.com/theislab/scCODA
They use bayesian inference to model differential expression from a CoDA perspective in single-celled datasets. Not sure if you've come across it but thought that it could be useful since you guys are currently developing BIRDMAn.
scCODAis built ontensorflow 2andtensorflow probability.https://www.nature.com/articles/s41467-021-27150-6
https://github.com/theislab/scCODA