National Institutes of Health (NIH), NIMH
Funded Project: A comprehensive whole-brain atlas of transcriptomic cell types in the mouse
Lead PI: Hongkui Zeng
Grant details: 1U19MH114830 | Sep 1, 2017 – Jun 30, 2022
Metazoan organs, including the brain, are composed of various cell types, but the cell type census for even a relatively simple mammalian brain such as the one of the mouse is not available. A whole mouse brain database of molecularly defined cell types organized in a taxonomy and annotated with their precise locations within the brain, would be of unprecedented importance for understanding the functions of mammalian brains. Recent technological advancements have enabled transcriptomic characterization of large numbers of individual cells through single cell RNA-sequencing (scRNA-seq). However, a number of methodologies exist, and differences in experimental and analysis methods employed by different labs, animal ages and strains, and spatial resolutions of dissections, make data comparison and integration challenging. In order to create a brain-wide atlas of cell types in the mouse as a standard for the field, we propose to utilize two scRNA-seq technologies at different scales and sequencing depths in a standardized manner across the entire mouse brain, to achieve consistency and comparability across different brain regions. We will compare these two methods, Smart-seq and droplet-based sequencing, to systematically assess to what extent they can distinguish different cell types. We will focus on the non-multiplexed, deeper sequencing method, Smart-seq, to generate a high-quality, foundational single-cell transcriptomic cell type atlas for all brain regions that are also precisely dissected and registered into a mouse brain common coordinate framework so that the spatial origin of each cell or cell type can be easily visualized. We will use RNA fluorescence in situ hybridization (FISH) to confirm expression patterns of cell type-specific gene markers derived from scRNA-seq data and further delineate the anatomical specificity of different cell types. Finally, we will obtain Smart-seq data from a selected set of retrogradely labeled neurons to define the correspondence between transcriptomic cell types and their connectional specificity. We will employ various computational approaches for clustering analysis of the large scRNA-seq datasets, to create a taxonomy of cell types within the whole mouse brain. This dataset will be useful for the entire community to mine and to compare with their own studies using a diverse range of existing and future single cell techniques.