Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

v1.1.0 #98

Merged
merged 27 commits into from
Apr 12, 2024
Merged

v1.1.0 #98

merged 27 commits into from
Apr 12, 2024

Conversation

dbdimitrov
Copy link
Collaborator

1.1.0 (12.04.2024)

  • Added a check for the subset of cell types in li.multi.dea_to_lr. Related to Add a check that for the subset of cell types in li.multi.dea_to_lr #92.

  • Split Local and Global Bivariate metrics. Specifically, I reworked completely the underlying code, though the API should remain relatively unchanged. With the exceptions of: 1) lr_bivar is now removed and bivar has been renamed to bivariate. This allowed me to remove a lot of redundancies between the two functions. 2) nz_threshold has been renamed to nz_prop for consistency with expr_prop in the remainder of the package. Related to Split global from local scores #44.

  • li.mt.bivariate parameter mod_added has been renamed to key_added due to this now refer to both .obsm and .mod - depedening whether an AnnData or MuData object is passed.

  • Added Global Lee's statistic, along with a note on weighted product that upon z-scaling it is equivalent to Lee's local statistic.

  • The Global L statistic and Global Moran's R are themselves basically identical. See Eq.22 from Lee and Eq.1 in Supps of SpatialDM.

  • Changed the li.mt.bivar parameter function_name to local_name for consistency and to avoid ambiguity with the newly-added global_name parameter.

  • Added bumpversion to manage versioning. Related to Add bumpversion + metadata #73.

  • Added max_runs and stable_runs parameters to enable the inference of robust causal networks with CORNETO. Related to Generate Union of Causal Net Predictions #82.

  • Optimized MISTy such that the matrix multiplication by weights is done only once, rather than for each target. Users can now obtain the weighted matrix via the misty.get_weighted_matrix function.

  • MISTy models are now passed externally, rather than being hardcoded. This allows for more flexibility in the models used. As an example, I also added a RobustLinearModel from statsmodels. Related to External + GPU-enabled MISTy models #74.

  • Removed forced conversion to sparse csr_matrix matrices in MISTy. Related to By default do not convert views to csr_matrix when building MistyData. #57.

@dbdimitrov dbdimitrov merged commit 1bd703b into main Apr 12, 2024
3 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant