v0.5.0

What's Changed

This version marks a shift towards formulaic designs, i.e. formula like "~condition + treatment", "~condition + condition:treatment", "~condition + exp(cofactor)", etc.

BREAKING CHANGE

python 3.9 is no longer supported.

DeseqDataSet

• Designs should now be provided to DeseqDataSet using the design argument, either in the form of a string representing a formulaic formula (e.g. "~condition + treatment", "~condition + condition:treatment", "~condition + exp(cofactor)"...), or an ndarray directly corresponding to a design matrix.
• design_factors is still supported but throws a DeprecationWarning
• continuous_factors is deprecated, as continuous type inference is handled by formulaic
• ref_level is deprecated
• Due to new decorated methods, DeseqDataSet is no longer picklable. A to_picklable_anndata() method was added to allow users to pickle results for later use.

DeseqStats

• Default contrasts are no longer supported, as they lead to too many errors
• Contrasts may be provided as before for categorical variables (e.g. ["treatment", "test", "control"]), or directly in the form of a contrast vector (a numpy array).
• For now, contrasts for continuous variables are directly specified with a contrast vector.
• lfc_shrink no longer supports a default coef argument

Changelog

• ENH Generic design support using formulaic by @BorisMuzellec in #328
• MAINT Deprecated pandas.DataFrameGroupBy.grouper by @BorisMuzellec in #332
• ENH add fit size factors type as an argument of DeseqDataSet by @regloeb in #345
• ENH Switch to formulaic-contrasts by @grst in #344
• BUG nbinomGLM returns before convergence by @BorisMuzellec in #349
• REF change type hints to PEP585 by @BorisMuzellec in #351
• ENH add access to control_genes as a DeseqDataSet attribute by @BorisMuzellec in #356

New Contributors

• @regloeb made their first contribution in #345
• @grst made their first contribution in #344

Full Changelog: v0.4.12...v0.5.0