A single-cell and spatial RNA-seq database for Alzheimer's disease (ssREAD)

Abstract

Alzheimer's Disease (AD) pathology has been increasingly explored through single-cell and single-nucleus RNA-sequencing (scRNA-seq & snRNA-seq) and spatial transcriptomics (ST). However, the surge in data demands a comprehensive, user-friendly repository. Addressing this, we introduce a single-cell and spatial RNA-seq database for Alzheimer's disease (ssREAD). It offers a broader spectrum of AD-related datasets, an optimized analytical pipeline, and improved usability. The database encompasses 1,053 samples (277 integrated datasets) from 67 AD-related scRNA-seq & snRNA-seq studies, totaling 7,332,202 cells. Additionally, it archives 381 ST datasets from 18 human and mouse brain studies. Each dataset is annotated with details such as species, gender, brain region, disease/control status, age, and AD Braak stages. ssREAD also provides an analysis suite for cell clustering, identification of differentially expressed and spatially variable genes, cell-type-specific marker genes and regulons, and spot deconvolution for integrative analysis. ssREAD is freely available at https://bmblx.bmi.osumc.edu/ssread/ .

Fig. 1. ssREAD Data characteristics and statistics (as of December 2023).

a–d Barplots show the number of datasets by species, condition, sex, and brain regions, respectively. e Treemap presents the breakdown of technologies deployed in mouse and human studies. f The comparative table showcases the number of studies and datasets across extant AD databases and data collection sources.

Fig. 2. Overview of ssREAD functions.

a Features and visual representations related to sc/snRNA-seq, encompassing cell type annotations, marker gene expressions, and graphic depictions via bar and violin plots. Each box showcases the minimum, first quartile, median, third quartile, and maximum average expression values of a cell type (Ast: n = 208, Endo: n = 28, EN: n = 2,696, IN: n = 740, MIC: n = 266, OPC: n = 195, and Olig: n = 1,546). Dots represent outliers. b Functions and visualizations pertinent to ST, highlighting H&E imagery, layer annotations, marker gene expressions, and spot deconvolutions. c DEG analysis, with comparisons drawn between categories like Cell Type 1 vs. Cell Type 2 (CT1 vs. CT2), AD vs. Control, Male vs. Female, Brain Regions 1 vs. 2, etc. p-values were calculated based on two-sided Wilcoxon Rank–sum test and adjusted using Bonferroni correction. d Functional enrichment analysis focusing on GO ontology and KEGG pathways. p-values were calculated using the Hypergeometric test from Enrichr and were adjusted using the Benjamini–Hochberg correction method. e Predictions of cell type-specific regulons. The following abbreviations are used for cell types: Ast astrocytes, Endo endothelial cells, EN excitatory neurons, IN inhibitory neurons, Mic microglia, Olig Oligodendrocytes, OPC oligodendrocyte precursor cells.

Fig. 3. Multi-dimensional analysis of spatially-informed sub-populations.

a Annotation of the six cortical layers alongside the adjacent white matter within two human middle temporal gyrus (MTG) brain samples (ST01101 and ST01103). b Detection of spatial domains by RESEPT. c Visualization using MAPLE to elucidate shared or unique spatial domains identified across the two Spatial Transcriptomics samples (ST01101 and ST01103). d Alluvial diagrams showcasing the progression of cells: originating from individual samples, aggregating into joint subpopulations, and culminating in layer annotations. e A heatmap depicting genes specific to the MAPLE-derived clusters for both AD and Control samples. f Heatmap representing the top 10 upregulated and top 10 downregulated genes distinguishing AD from Control within Cluster 1. g Gene Set Enrichment Analysis (GSEA) of DEGs from (F) plotted against REACTOME pathways. The bar plot shows the top 10 upregulated and downregulated pathways, accompanied by normalized enrichment scores. h Spatial feature plots highlighting the variance in gene expression of PLP1 and UCHL1 from Cluster 1, segregated by AD and Control samples. i Violin plots showcasing the activity of two selected TFs between AD and Control, with associated p-values calculated from a two-sided Wilcoxon rank-sum test. Each box showcases the minimum, first quartile, median, third quartile, and maximum ARI results of a tool performed on different data subsets (Control group: n = 232, and AD group: n = 1412). Dots represent spatial spots. Source data are provided as a Source Data file.

Fig. 4. ssREAD facilitates comprehensive spatial transcriptomics analyses synergized with scRNA-seq datasets.

a UMAP representation of cell types derived from the Seattle Alzheimer’s Disease Brain Cell Atlas (SEA-AD). Clockwise from the top left, corner insets elucidate the Braak stage, Thal phase, ethnicity, and sex attributes. b–e Bar plots visualizing the distribution of cells based on the Braak stage, Thal phase, sex, and ethnicity, categorized by the condition in the atlas. f Bar plots displaying the fractional representation of cell types, contrasting AD and control within the SEA-AD atlas.  g Heatmap of top and bottom 25 DEGs identified between AD and control samples in the integrated AD035 MTG dataset in Microglia. h Comparison of DEGs overlap in Microglia among AD035 MTG, AD048 PFC, and their integrated datasets. i–l Deconvolution analysis of cell types between ST samples (ST01101 and ST01103) and the scRNA-seq SEA-AD atlas, showing the cell type fractions for Astrocytes (j) and Oligodendrocytes (l). Marker gene expression indicators, GFAP for Astrocytes (i) and MOBP for Oligodendrocytes (k). m–t Spatial distributions of DEGs in AD and control samples within MAPLE cluster 1, cross-validated with single-cell DEGs from the integrated AD035 datasets. Upregulated DEGs in Astrocytes GJA1 (m) and MT-ATP8 (n). Downregulated DEGs in Astrocytes IFITM3 (q) and TUBB2B (r). Upregulated DEGs in Oligodendrocyte ERBIN (o), GPRC5B (p), MID1IP1 (s), and SLC44A1 (t).

Fig. 5. Exploration of sex-specific differences at the cellular level in AD.

a UMAP visualization of the single-cell data used in the analysis, with different cell types color-coded. b–c Bar plots illustrate the count and proportion of each cell type, segregated by sex. This reveals any potential differences in cellular composition between male and female samples. d UpSet plot showing the unique and shared DEGs across four groups: Male AD patients, Female AD patients, Male controls, and Female controls. e Violin plots for the top 10 upregulated DEGs between male and female in microglia. * Indicates sex-chromosomal genes. p-values were calculated based on a two-sided Wilcoxon Rank-Sum test and adjusted using Bonferroni correction. f Violin plots for the top 10 downregulated DEGs between male and female in microglia. p-values were calculated based on a two-sided Wilcoxon Rank-Sum test and adjusted using Bonferroni correction. g Gene Set Enrichment Analysis (GSEA) plot showing the enrichment of genes involved in binding and uptake of ligands by scavenger receptors. h–j GSEA plots showing the enrichment of genes involved in neurogenesis for three different comparisons: Male AD vs. Male Control in Microglia (h), Female AD vs. Female Control in Microglia. (i), and Male AD vs. Female AD in Microglia (j). These plots highlight the sex-specific differences in neurogenesis-related gene activity under AD conditions.