Preprocessed Datasets for scMoAnno: Cell Type Annotation with Paired Single-Cell Multi-Omics Data

This repository contains preprocessed datasets used in scMoAnno, a deep learning framework designed for accurate cell type annotation using paired single-cell multi-omics data (scRNA-seq and scATAC-seq). scMoAnno addresses the limitations of single-omics approaches and improves generalization, particularly in the identification of rare cell types, through a two-stage supervised learning strategy. The data are packaged in a .7z archive and include processed versions of the following public datasets: 10X Multiome ISSAAC-seq SHARE-seq Multiome-BMMC These datasets have been preprocessed to be directly usable with our framework. The original raw data can be accessed from the following sources: Box (UPenn) ArrayExpress: E-MTAB-11264 We adhere to the data usage policies of the original sources and make our processed data available for the benefit of the research community, supporting reproducibility and further studies in multi-omics integration and cell type annotation.