We present a machine-learning method to prioritize the cell types most responsive to biological perturbations within highdimensional single-cell data. We validate our method, Augur (https://github.com/neurorestore/Augur), on a compendium of single-cell RNA-seq, chromatin accessibility, and imaging transcriptomics datasets. We apply Augur to expose the neural circuits that enable walking after paralysis in response to spinal cord neurostimulation. a, Schematic overview of Augur. b, AUCs of Augur and a naive random forest classifier without subsampling in simulated scRNA-seq datasets containing increasing numbers of cells. Cell type prioritizations are confounded by training dataset size for the naive classifier, but Augur abolishes this confounding factor. The mean and standard deviation of ten simulation replicates are shown. c, Pearson correlations between the AUC of each cell type, and the number of cells of that type sequenced, across a compendium of 22 scRNA-seq datasets, for Augur and a naive random forest classifier without subsampling. d, Augur AUCs scale monotonically with both the proportion of DE genes and the magnitude of DE in simulated cell populations. e, Relationship between number of DE genes detected by a representative test for single-cell differential gene expression (Wilcoxon rank-sum test), and the proportion of differentially expressed genes simulated between the two populations, for simulated populations of between 100 and 1,000 cells. f, Augur cell type prioritizations track with duration of LPS exposure in a cross-species scRNA-seq experiment 7. Grey points show AUCs with sample labels randomly permuted. g-h, Cell type prioritization in matched single-cell 4 and bulk 8 transcriptomic profiles of PBMCs after interferon stimulation. g, Left, Augur cell type prioritizations mirror the number of DE genes in a microarray dataset of FACS-purified cells. Right, the number of DE genes detected in the scRNA-seq dataset by a Wilcoxon rank-sum test is uncorrelated with the FACS gold standard. h, Correlation coefficients between cell type prioritizations (AUC or number of DE genes) in the scRNA-seq dataset and the FACS gold standard. i-j, Cell type prioritization in the mouse ventromedial hypothalamus reflects induction of IEG transcription. i, Correlation between AUC and the difference in the first principal component of IEG expression (∆IEG eigengene) engaging in aggressive behavior. j, Pearson correlation coefficients between cell type-specific AUC and ∆IEG eigengene values for eleven behavioral stimuli 9. k, Reproducibility of cell type prioritization in two independent scRNA-seq studies of Alzheimer's disease 5,10. l, Augur cell type prioritizations in a scATAC-seq dataset 11 track with the number of DE genes in an RNA-seq dataset of FACS-purified cells.
jordan.squair@epfl.ch a, Schematic overview of Augur. b, AUCs of Augur and a naive random forest classifier without subsampling in simulated scRNA-seq datasets containing increasing numbers of cells. Cell type prioritizations are confounded by training dataset size for the naive classifier, but Augur abolishes this confounding factor. The mean and standard deviation of ten simulation replicates are shown. c, Pearson correlations between the AUC of each cell type, and the number of cells of that type sequenced, across a compendium of 22 scRNA-seq datasets, for Augur and a naive random forest classifier without subsampling. d, Augur AUCs scale monotonically with both the proportion of DE genes and the magnitude of DE in simulated cell populations. e, Relationship between number of DE genes detected by a representative test for single-cell differential gene expression (Wilcoxon rank-sum test), and the proportion of differentially expressed genes simulated between the two populations, for simulated populations of between 100 and 1,000 cells. f, Augur cell type prioritizations track with duration of LPS exposure in a cross-species scRNA-seq experiment 7 . Grey points show AUCs with sample labels randomly permuted. g-h, Cell type prioritization in matched single-cell 4 and bulk 8 transcriptomic profiles of PBMCs after interferon stimulation. g, Left, Augur cell type prioritizations mirror the number of DE genes in a microarray dataset of FACS-purified cells. Right, the number of DE genes detected in the scRNA-seq dataset by a Wilcoxon rank-sum test is uncorrelated with the FACS gold standard. h, Correlation coefficients between cell type prioritizations (AUC or number of DE genes) in the scRNA-seq dataset and the FACS gold standard. i-j, Cell type prioritization in the mouse ventromedial hypothalamus reflects induction of IEG transcription. i, Correlation between AUC and the difference in the first principal component of IEG expression (∆IEG eigengene) engaging in aggressive behavior. j, Pearson correlation coefficients between cell type-specific AUC and ∆IEG eigengene values for eleven behavioral stimuli 9 . k, Reproducibility of cell type prioritization in two independent scRNA-seq studies of Alzheimer's disease 5,10 . l, Augur cell type prioritizations in a scATAC-seq dataset 11 track with the number of DE genes in an RNA-seq dataset of FACS-purified cells.
Providing a common place for the civil society to gather and discuss topics of mutual interest is a growing challenge for social and collaborative computing. Web-based tools for civic engagement, while promising, are still disconnected from meaningful physical locations where citizens usually meet and might limit the involvement of a considerable portion of the citizen population. We propose a system, Agora2.0, designed to recover the useful function that public places have had in the past in promoting and regulating citizens' participation in public decisions. Agora2.0 is inspired by the old concept of the Greek agora, or public square. It is composed of an onsite interactive public display and an online site. We present the project, the analysis of the requirements, the system prototype, and its evaluation during deployments in a university and in a public relations office of a European city.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.