Add linear mixed effects model to DE interface
Open
Nobody has claimed this yet.
enhancement
- Dominant language
- Python
- Stars
- 345
- Forks
- 66
- Avg merge
- 1d 4h
- Merged PRs (30d)
- 13
Description
Description of feature
Description of feature
They can be used to either perform tests on all cells, including the sample as random effect, or on pseudobulk when there are designs that require random effects.
Methods
Model specification
- Neither formulaic nor patsy currently allow for random effects. There's a PR open to add support https://github.com/matthewwardrop/formulaic/pull/34
- formulae allows specifying random effects: https://github.com/bambinos/formulae
- Alternatively, it could be possible to specify two formulas, one for fixed and one for random effects.
Migrated from https://github.com/scverse/multi-condition-comparisions/issues/20
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
The issue names the DE interface but no repository files or tests; start by locating that interface and reviewing the linked MAST, DREAM, Statsmodels LME, formulaic, and formulae references. Resolve how fixed and random effects should be specified, then define tests covering cell-level and pseudobulk designs with random effects before implementing the feature.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- bioinformatics
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100