A SPATIAL TRANSCRIPTOMICS ATLAS OF THE ADULT MOUSE BRAIN

Understanding the spatial organization of cell types in the brain is essential for deciphering its intricate functional and anatomical architecture. Spatial transcriptomics offers a method to correlate transcriptomic cell types with their anatomical locations. Recently, we published a whole-brain multiplexed error-robust fluorescence in situ hybridization (MERFISH) dataset (Yao et al., 2023), which includes 4.3 million cells, complementing our single-cell RNA sequencing (scRNAseq) taxonomy of 5322 clusters. This spatial dataset allowed us to examine the distribution of cell types within the mouse brain.
Following this achievement, we expanded our datasets. Our study encompasses six adult datasets, comprising four P56 and two P28 samples, all sex-matched to facilitate robust comparative analyses. A novel segmentation and processing pipeline was employed, resulting in a dataset containing approximately 33 million cells, registered to the Common Coordinate Framework (CCFv3) for uniform spatial alignment and comparative studies. This comprehensive dataset provides insights into brain region relationships based on cell type composition and enables analyses of cell type distributions across sexes with enhanced statistical power.
Additionally, employing the transformer-based spatial domain detection tool, CellTransformer, we generated computationally derived brain parcellations for detailed spatial mapping. These datasets represent a valuable resource for advancing the understanding of cell type composition and spatial organization within the mammalian brain.