Package: gedi2 2.3.6

gedi2: Gene Expression Decomposition and Integration

A memory-efficient implementation for integrating gene expression data from single-cell RNA sequencing experiments. Uses a C++ backend with thin R wrappers to enable analysis of large-scale single-cell datasets. The package supports multiple data modalities including count matrices, paired data (splicing, RNA velocity, CITE-seq), and binary indicators. It implements a latent variable model with block coordinate descent optimization for dimensionality reduction and batch effect correction. The method is described in Mikaeili Namini et al. (2026) <doi:10.1093/bioinformatics/btag334>, building on the original 'GEDI' model of Madrigal et al. (2024) <doi:10.1038/s41467-024-50963-0>.

Authors:Arsham Mikaeili Namini [aut, cre], Hamed S.Najafabadi [aut]