incompatibility with moscot due to ott-jax dependency

[cvduffy]

The latest releases of moscot and cellflow are incompatible due to conflicting ott-jax version pins.

Will cellflow be refactored to accommodate the new ott-jax flow matching module à la theislab/moscot#839?

In the meantime, are there any suggestions for working with the latest moscot and cellflow releases concurrently?

Thanks!

[MUCDK]

Hi @cvduffy ,

Yes, we will update it soon. @selmanozleyen @lorenzo-consoli could you please coordinate?
I don't think there is a way to do it in the same environment. I would just use the output of one, and then continue with the other package in a new environment. Out of curiosity, for which use case do you require both?

[cvduffy]

Thank you! Use case is exploratory - I'm using CellFlow to predict perturbation outcomes but I think moscot is better suited for describing/modeling single-cell data I've already collected. For example to get OT pairings between control and perturbed cells, moscot exposes a lot of options that I don't think are straightforwardly accessible via CellFlow. Let me know if that sounds reasonable / if I'm missing something!

[selmanozleyen]

I will be on it