Systematic evaluation of single-cell multimodal data integration for comprehensive human reference atlas

作者：Mario Acera-Mateos, Xian Adiconis, Jessica-Kanglin Li, Doménica Marchese, Ginevra Caratù, Chung-Chau Hon, Prabha Tiwari, Miki Kojima, Beate Vieth, Michael Murphy, Sean Simmons, Thomas Lefévre, Irene Claes, Christopher L. O’Connor, Rajasree Menon, Edgar A. Otto, Yoshinari Ando, Katy Vandereyken, Matthias Kretzler, Markus Bitzer, Ernest Fraenkel, Thierry Voet, Wolfgang Enard, Piero Carninci, Holger Heyn, Joshua Z. Levin, Elisabetta Mereu · 发表于：bioRxiv (Cold Spring Harbor Laboratory) · 年份：2025 · DOI：10.1101/2025.03.06.637075 · 被引用次数：4 · 研究领域：Single-cell and spatial transcriptomics、Context-Aware Activity Recognition Systems

The integration of multimodal single-cell data enables comprehensive organ reference atlases, yet its impact remains largely unexplored, particularly in complex tissues. We generated a benchmarking dataset for the renal cortex by integrating 3' and 5' scRNA-seq with joint snRNA-seq and snATAC-seq, profiling 119,744 high-quality nuclei/cells from 19 donors. To align cell identities and enable consistent comparisons, we developed the interpretable machine learning tool scOMM (single-cell Omics Multimodal Mapping) and systematically assessed integration strategies. "Horizontal" integration of scRNA and snRNA-seq improved cell-type identification, while "vertical" integration of snRNA-seq and snATAC-seq had an additive effect, enhancing resolution in homogeneous populations and difficult-to-identify states. Global integration was especially effective in identifying adaptive states and rare cell types, including WFDC2-expressing Thick Ascending Limb and Norn cells, previously undetected in kidney atlases. Our work establishes a robust framework for multimodal reference atlas generation, advancing single-cell analysis and extending its applicability to diverse tissues.