Type 2 diabetes Reprograms Bone Marrow Hematopoiesis and Dysregulates Immune Signaling in Response to Stroke

Data Repository for scRNA-seq, GeoMx digital spatial profiling (DSP) and Nanostring nCounter Analysis in the bone marrow in T2DM Stroke Model This repository contains the raw transcriptomic data and Seurat objects generated for analyzing hematopoiesis and immune cell regulation of bone marrow in non-diabetes (db/+) and diabetes (db/db) after ischemic stroke. Data was generated using single-cell RNA sequencing (scRNA-seq) for bone marrow cell annotation, cell-cell communication and gene expression. Geomx DSP and nCounter analysis validated the gene expression in monocytes and neutrophils in the bone marrow. This study aims to reveal how diabetes reprograms bone marrow hematopoiesis after ischemic stroke, driving maladaptive immune regulation. Repository Structure Root Files README.md: This file, describing the repository and data files. sc RNA seq_raw_data: Raw scRNA-seq data files. GeoMx-analysis_BM_raw data: Raw GeoMx DSP data files. ncounter_analysis-BM_raw data: Raw nCounter data files. Seurat object: Contains Seurat objects for clustering and visualization of whole bone marrow. CellChat object: Contains RDA files for CellChat analysis. File Descriptions sc RNA seq_raw_data/ dbdb-MCAO.zip: Raw scRNA-seq data for db/db mice post-distal middle cerebral artery occlusion (dbdb.Stroke). dbdb-sham.zip: Raw scRNA-seq data for db/db mice with sham surgery (dbdb). db_pos-sham.zip: Raw scRNA-seq data for db/+ mice with sham surgery (control). db_pos-MCAO.zip: Raw scRNA-seq data for db/+ mice post-DMCAO (db+. Stroke). Methods: scRNA-seq was performed using 10X Genomics GemCode Technology. Data were processed with Cell Ranger (v1.3), and differential gene expression analysis was conducted in Seurat with normalization based on UMI counts. Seurat object/ bonemarrow_resolved.rda: Seurat object containing processed scRNA-seq data of bone marrow cells from both diabetic and normoglycemic mice under stroke and sham conditions. Contains cell clusters annotated using SingleR and the Tambula Muris database. Methods: Filtering was performed for cells with fewer than 500 detected genes, and data were normalized using log-normalization. Clustering was carried out with PCA and visualized with UMAP. CellChat object/ T2DM.Stroke_cellchat.rda: CellChat object for the db/db mice with ischemic stroke. Analysis was conducted to identify cell-cell communication patterns altered due to stroke in diabetic conditions. T2DM_cellchat.rda: CellChat object for db/db control (no stroke) mice. Provides baseline data for diabetic conditions. Stroke_cellchat.rda: CellChat object for the db/+ (normoglycemic) mice with ischemic stroke, representing the stroke model in non-diabetic conditions. Ctrl_cellchat.rda: CellChat object for db/+ control (no stroke) mice. Baseline for non-diabetic, non-stroke conditions. Methods: CellChat analysis identifies ligand-receptor interactions to reveal cross-talk between bone marrow cells, especially focusing on hematopoietic precursor cells (HPC1) and monocytes. These data provide insights into immune dysregulation through identified signaling pathway activation. GeoMx-analysis_BM_raw data/ annotation: annotation files contain metadata describing the experimental design and spatial regions that were analyzed, including sample identifiers, region of interest (ROI) labels, cell segment, experimental and group conditions, and other relevant sample-level annotations. dccs: DCC files contain the raw digital counts for each target measured by the GeoMx DSP. Each DCC file corresponds to a single region of interest and reports the number of detected barcode counts for each gene or protein target prior to normalization. pkcs: PKC files contain information related to probe performance and quality control, including counts for individual probes targeting the same gene or protein. Methods: GeoMx samples were sequenced on an Illumina NovaSeq 6000 platform. GeoMx RNA expression data were analyzed in R using the GeomxTools package for quality control, filtering, normalization, and dimensionality reduction with UMAP. The GeoMx object was then converted into a Seurat object for differential gene expression and Gene Ontology (GO) enrichment analysis. ncounter_analysis-BM_raw data/ .RCC: .RCC files are the primary raw data output generated by the NanoString nCounter system. Each RCC file corresponds to a single biological sample and contains the digital counts for all targets measured in the assay. Methods: Gene expression profiling was performed using the NanoString nCounter® Myeloid Innate Immunity Panel v2. Data normalization and pathway analysis were conducted using nSolver 4.0 software. Gene expression boxplots were generated in R. License Data and scripts are available for non-commercial use under a Creative Commons License.