Search

1AbstractAccurate identification of cell classes across the tissues of living organisms is central in the analysis of growing atlases of single-cell RNA sequencing (scRNA-seq) data across biomedicine. Such analyses are often based on the existence of highly discriminating “marker genes” for specific cell classes which enables a deeper functional understanding of these classes as well as their identification in new, related datasets. Currently, marker genes are defined by methods that serially assess the level of differential expression (DE) of individual genes across landscapes of diverse cells. This serial approach has been extremely useful, but is limited because it ignores possible redundancy or complementarity across genes, that can only be captured by analyzing several genes at the same time. We wish to identify discriminatingpanelsof genes. To efficiently explore the vast space of possible marker panels, leverage the large number of cells often sequenced, and overcome zero-inflation in scRNA-seq data, we propose viewing panel selection as a variation of the “minimal set-covering problem” in combinatorial optimization which can be solved with integer programming. In this formulation, the covering elements are genes, and the objects to be covered are cells of a particular class, where a cell is covered by a gene if that gene is expressed in that cell. Our method, CellCover, identifies a panel of marker genes in scRNA-seq data that covers one class of cells within a population. We apply this method to generate covering marker gene panels which characterize cells of the developing mouse neocortex as postmitotic neurons are generated from neural progenitor cells (NPCs). We show that CellCover captures cell class-specific signals distinct from those defined by DE methods and that CellCover’s compact gene panels can be expanded to explore cell type specific function.Transfer learning experiments exploring these covering panels acrossin vivomouse, primate, and human scRNA-seq datasets demonstrate that CellCover identifies markers of conserved cell classes in neurogenesis, as well as markers of temporal progression in the molecular identity of these cell types across development of the mammalian neocortex. The gene covering panels we identify across cell types and developmental time can be freely explored in visualizations across all the public data we use in this report at with NeMo Analytics [1] throughhttps://nemoanalytics.org/p?l=CellCover. The code for CellCover is written in R and the Gurobi R interface and is available at [2].Graphical Abstract

MotivationIt is necessary to develop exploratory tools to learn from the unprecedented volume of high-dimensional multi-omic data currently being produced. We have developed an R package, SJD, which identifies components of variation that are shared across multiple matrices. The approach focuses specifically on variation across the samples/cells within each dataset while incorporating biologist-defined hierarchical structure among input experiments that can spanin vivoandin vitrosystems, multi-omic data modalities, and species.ResultsSJD enables the definition of molecular variation that is conserved across systems, those that are shared within subsets of studies, and elements unique to individual matrices. We have included functions to simplify the construction and visualization of highly complexin silicoexperiments involving many diverse multi-omic matrices from multiple species. Here we apply SJD to decompose four RNA-seq experiments focused on neurogenesis in the neocortex. The public datasets used in this analysis and the conserved transcriptomic dynamics in mammalian neurogenesis that we define here can view viewed and explored together athttps://nemoanalytics.org/p?l=ChenEtAlSJD2022&g=DCX.Availability and ImplementationThe SJD package can be found athttps://chuansite.github.io/SJD.Contacthzchenhuan@gmail.com;ccolant1@jhmi.eduSupplementary informationSupplementary data are available at Bioinformatics online.

Scalable technologies to sequence the transcriptomes and epigenomes of single cells are transforming our understanding of cell types and cell states. The Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative Cell Census Network (BICCN) is applying these technologies at unprecedented scale to map the cell types in the mammalian brain. In an effort to increase data FAIRness (Findable, Accessible, Interoperable, Reusable), the NIH has established repositories to make data generated by the BICCN and related BRAIN Initiative projects accessible to the broader research community. Here, we describe the Neuroscience Multi-Omic Archive (NeMO Archive; nemoarchive.org), which serves as the primary repository for genomics data from the BRAIN Initiative. Working closely with other BRAIN Initiative researchers, we have organized these data into a continually expanding, curated repository, which contains transcriptomic and epigenomic data from over 50 million brain cells, including single-cell genomic data from all of the major regions of the adult and prenatal human and mouse brains, as well as substantial single-cell genomic data from non-human primates. We make available several tools for accessing these data, including a searchable web portal, a cloud-computing interface for large-scale data processing (implemented on Terra, terra.bio), and a visualization and analysis platform, NeMO Analytics (nemoanalytics.org).

Scalable technologies to sequence the transcriptomes and epigenomes of single cells are transforming our understanding of cell types and cell states. The Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative Cell Census Network (BICCN) is applying these technologies at unprecedented scale to map the cell types in the mammalian brain. In an effort to increase data FAIRness (Findable, Accessible, Interoperable, Reusable), the NIH has established repositories to make data generated by the BICCN and related BRAIN Initiative projects accessible to the broader research community. Here, we describe the Neuroscience Multi-Omic Archive (NeMO Archive; nemoarchive.org), which serves as the primary repository for genomics data from the BRAIN Initiative. Working closely with other BRAIN Initiative researchers, we have organized these data into a continually expanding, curated repository, which contains transcriptomic and epigenomic data from over 50 million brain cells, including single-cell genomic data from all of the major regions of the adult and prenatal human and mouse brains, as well as substantial single-cell genomic data from non-human primates. We make available several tools for accessing these data, including a searchable web portal, a cloud-computing interface for large-scale data processing (implemented on Terra, terra.bio), and a visualization and analysis platform, NeMO Analytics (nemoanalytics.org).KEY POINTSThe Neuroscience Multi-Omic Archive serves as the genomics data repository for the BRAIN Initiative.Genomic data from >50 million cells span all the major regions of the brains of humans and mice.We provide a searchable web portal, a cloud-computing interface, and a data visualization platform.

Inflammation early in life is a clinically established risk factor for autism spectrum disorders and schizophrenia, yet the impact of inflammation on human brain development is poorly understood. The cerebellum undergoes protracted postnatal maturation, making it especially susceptible to perturbations contributing to risk of neurodevelopmental disorders. Here, using single-cell genomics, we characterize the postnatal development of cerebellar neurons and glia in 1-5-year-old children, comparing those who died while experiencing inflammation vs. non-inflamed controls. Our analyses reveal that inflammation and postnatal maturation are associated with extensive, overlapping transcriptional changes primarily in two subtypes of inhibitory neurons: Purkinje neurons and Golgi neurons. Immunohistochemical analysis of a subset of these brains revealed no change to Purkinje neuron soma size but evidence for increased activation of microglia in those subjects experiencing inflammation. Maturation- and inflammation-associated genes were strongly enriched for those implicated in neurodevelopmental disorders. A gene regulatory network model integrating cell type-specific gene expression and chromatin accessibility identified seven temporally specific gene networks in Purkinje neurons and suggested that the effects of inflammation correspond to blunted cellular maturation.One Sentence SummaryPost-mortem cerebelli from children who perished under conditions that included inflammation exhibit transcriptomic changes consistent with blunted maturation of Purkinje neurons compared to those who succumbed to sudden accidental death.