Additional file 1 of MSCsDB: a database of single-cell transcriptomic profiles and in-depth comprehensive analyses of human mesenchymal stem cells

Additional file1: Figure S1. The information on MSC atlas taxonomy. (A) UMAP of all MSCs with cluster annotations, (B) UMAP of MSCs color-labelled by tissue, (C) Cell counts of MSCs from different tissues in each cluster, and (D) Cell counts of MSCs from different samples in each cluster. Figure S2. Differentiation scoring of MSCs on five differentiation directions. (A) Scoring of osteogenesis, chondrogenesis, adipogenesis, myogenesis and neurogenesis. (B) Scoring of representative gene expression for MSCs differentiation. Figure S3. Home page of MSCsDB. which includes website introduction, functionality overview, gene cloud, and website update news. Figure S4. Module of Dataset and link to the module of Explore. Users can view the metadata of each sample dataset, such as the original article, data repository and sequencing technology. Users can also click on the “Explore” button to view the sample’s clustering annotation, gene expression level analysis, pathway enrichment analysis, copy number variation analysis, and pseudotime analysis results. Figure S5. Functionality in the module of Atlas. (A) UMAP of MSCs with cluster annotations. Users can select specific clusters to view their distribution. The MSC atlas can also be classified by tissue or batch and shown separately. (B) Gene signature of MSCs. Users can analyze the cell percentage of all genes and click on the “View” button to view the gene expression levels in cells and clusters. The Gene Card database is also linked for users to view gene information. Users can also enter a specific gene in the search box to retrieve relevant information. Figure S6. An example of functionality in the module of Atlas. (A) Pathway enrichment analysis of MSCs from different databases. Users can switch between different databases. Users can also select specific clusters and pathways to view their enrichment status. (B) Copy number variation analysis of MSCs using copyKat and InferCNVpy packages. The copyKat software can predict whether the cells are normal cells (diploid) or tumor cells (aneuploid). The InferCNVpy package gives prediction values, so we provide chromosome heatmaps based on CNV clustering for users to distinguish between normal cells and tumor cells. (C) Pseudotime analysis of MSCs using PAGA method. We show the cell trajectory inference plot and cluster UMAP plot for a single sample. (D) Transcription factor network analysis of MSCs using pyscenic package. We provide the transcription factor network analysis result table and heatmap for a single sample’s cluster. Users can click on the “View” button in the table to view the target genes regulated by that transcription factor. Figure S7. De novo analysis for clustering, pathway enrichment, and quality evaluation. (A) UMAP plot of MSC clustering and annotation using Scanpy package for a sample dataset. (B) Pathway enrichment analysis using Clusterprofiler package for a sample dataset. (C) Copy number variation analysis using CopyKat and InferCNVpy packages for a sample dataset. Figure S8. De novo analysis for pseudotime and gene regulatory network analysis. (A) Pseudotime analysis using PAGA method for a sample dataset. (B) Gene regulatory network analysis using pyscenic package for a sample dataset. Table S1. Marker genes used for potency score analysis. Table S2. Scoring for each cluster using geneset.