Towards an integrated analytics solution to creating a spatially-resolved single-cell multi-omics brain atlas

点击翻译按钮获取中文摘要

PROJECT SUMMARY The maintenance and function of the nervous system depends on cell-cell interactions among neuronal and non-neuronal cell populations, which occur through physically binding cell membrane surface or secreted proteins, triggering signaling cascades that activate cell-type gene regulatory programs. The cell-cell interactome responds and regulates the microenvironment which is altered in physiological processes such as brain development and aging, or during the onset and progression of different neuropsychiatric disorders. Although great progress has been made in dissecting cellular heterogeneity through single-cell transcriptome and epigenome profiling, a spatial atlas deciphering the gene regulatory programs that mediate cell-cell interactions among neuronal and non-neuronal cell populations in the human nervous system is lacking. New technologies have been developed to generate transcriptome-scale gene expression profiles with high spatial resolution in the original tissue context, thus greatly facilitating the structural characterization of the tissue microenvironment as well as mechanistic investigation of cell-cell interactions. Spatial transcriptomics technologies are increasingly used by the NIH Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative and viewed as an important component of the cell atlas building effort. Integrative analyses of spatial transcriptomics and multi-omic data have great potential in constructing a complete spatially resolved brain atlas. However, it remains challenging to systematically analyze, compare, and integrate data generated by different groups, in part due to the significant variation in technology platforms and data analysis pipelines. In previous work, our lab has developed Giotto as a powerful toolbox for comprehensive spatial transcriptomics analysis. In this project, we propose to use Giotto as the basis to create a cloud-based software platform and further develop new pipelines to facilitate harmonizing and integrating BRAIN Initiative datasets. The specific aims are: 1. To develop a cloud-based software platform for harmonizing spatial transcriptomics data analysis and visualization; 2. To elucidate the gene regulatory networks mediating cell- cell interactions in the brain through integrated spatial multi-omic analysis; and 3. To create a benchmark system for objective evaluation of spatial transcriptomics technologies and data analysis pipelines. Taken together, our proposed research will generate powerful computational tools that will not only help BRAIN Initiative investigators to construct a spatially resolved brain atlas but also enable the broad community to efficiently utilize the rich resources generated by the BRAIN Initiative in their own research.