moscot
moscot copied to clipboard
theislab
Multi-omic single-cell optimal transport tools
there are two options: - use callback for graph compputation (with epsilon=4t) - use `set_graph_xy`, `set_graph_x`, `set_graph_y`
E.g. `pca-callback` and `spatial_norm_callback`.
``` sp = SinkhornProblem(adata_red) sp = sp.prepare( key="perturbation", joint_attr="X_pca", policy="star", reference="control" ) sp = sp.solve(1e-2, 0.95, 0.95) sp ``` generates `SinkhornProblem[('Dacinostat (LAQ824)', 'control'), ('Givinostat (ITF2357)', 'control'), ('TAK-901', 'control'), ('Flavopiridol HCl',...
All downstream methods should be linear in memory, i.e. allow for batch-wise computation. > What are the memory requirement for the mp.impute() and mp.correlate() functions? For mp.impute(), if I try...
Can we create an issue for this? _Originally posted by @michalk8 in https://github.com/theislab/moscot/pull/633#discussion_r1444077840_
In #633 , we currently only have tests for the `TemporalProblem`. As the pipeline for a graph-based cost is quite different from the remaining costs, we need tests for all...
Would be nice if moscot problems would have a `.copy` method implemented. That would make it easier to experiment with e.g. parameters on the same problem and compare results.
This is a feature request for methods like `score_genes_for_marginals`. It's generally bad practice to use gene symbols as the var_names for an AnnData object, since they're not unique and can...
