Metadata-Version: 2.4
Name: fedpylimma
Version: 0.0.1
Summary: Federated limma-voom for privacy-preserving differential expression analysis of bulk RNA-seq data (name reservation; full release forthcoming)
Author: Shukrullo Foziljonov
Project-URL: Homepage, https://github.com/byShukrullo/fedpylimma
Project-URL: Source, https://github.com/byShukrullo/fedpylimma
Project-URL: Issues, https://github.com/byShukrullo/fedpylimma/issues
Keywords: differential expression,RNA-seq,federated learning,limma,voom,bioinformatics,flower
Classifier: Development Status :: 1 - Planning
Classifier: Intended Audience :: Science/Research
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
Classifier: Operating System :: OS Independent
Requires-Python: >=3.10
Description-Content-Type: text/markdown

# fedpylimma

**Federated limma-voom for privacy-preserving differential expression analysis
of bulk RNA-seq data.**

fedpylimma extends the limma-voom differential expression workflow to the
federated setting, so that multiple institutions can jointly analyse RNA-seq
data without sharing raw counts. It is built on the [Flower](https://flower.ai)
federated learning framework and exploits the additive separability of the
weighted linear-model normal equations, so that server-side aggregation
reproduces the centralized estimate exactly.

> **Status: placeholder / name reservation.**
> This release reserves the project name. The full implementation accompanies
> the author's MSc thesis and will be published here on release.

## Planned features

- Federated limma-voom (upper-quartile normalization) on the Flower framework
- Multi-group comparisons (beyond two-group designs)
- Exact equivalence between federated and centralized results

## Author

Shukrullo Foziljonov, MSc Biotechnology and Applied AI for Health,
University of Pisa.
