A tumor profiling resource for ovarian cancer: insights into chemotherapy-driven heterogeneity and personalized treatment strategy: scRNA-seq, CyTOF and IMC single-cell data

This data was generated as part of the Tumor Profiler study. If you use it in your research, please cite: Jacob, F., Wegmann, R., Ficek-Pascual, J. et al:, A tumor profiling resource for ovarian cancer: insights into chemotherapy-driven heterogeneity and personalized treatment strategy. Nature Communications, 2026 Derived data - scRNA-seq This is a SingleCellExperiment object saved as an RDataSet. It contains the following slots: Assays: counts: raw counts colData: Cell-level metadata barcodes: The cell arcode fractionMT: Fraction mitochondrail genes per cell n_umi: Total number of UMIs per cell n_gene: Total number of genes per cell log_umi: log10 total number of UMIs per cell g2m_score: Cell cycle phase score for G2M s_score: Cell cycle phase score for S cycle_phase: predicted cell cycle phase celltype_major_full_ct_name: Major cell typ efull name celltype_major: Major cell type hort name celltype_final_full_ct_name: Cell subtype full name celltype_final: Cell subtype short name sample_id rowData: Gene-level metadata gene_ids gene_names Derived data - IMC This dataset contains all imaging mass cytometry (IMC) data stored as a SingleCellExperiment object. Each cell is annotated with XY spatial coordinates, basic cell type, and image/sample ID. It includes raw counts, compensated counts, and a compensated (arcsinh-transformed) expression assay for 41 markers. Metadata contains sample identifiers, panel information, and unique core IDs. Dimensionality reductions (UMAP and t-SNE) are provided for spatial and cellular analysis. Derived data - CyTOF This is an R data set (.RDS) containing a SingleCellExperiment object with the following slots: Assays: counts_raw: signal intensity based on CyTOF dual counts exprs_raw: arcsinh transformed raw counts (cofactor 5) counts: batch corrected raw counts (linear scaling based on a quantile) exprs: arcsin transformed counts (cofactor 5) scaled: 0-1 normalized exprs (clipped to the 99.95th percentile) colData (cell metadata) bc_id: barcode of the sample during staining type: Sample type (ascites or tumor tissue) sample_id: TuPro sample ID pred_id: Predicted cell type [char] pred_n: Predicted cell type [integer] rowData (marker metadata) channel_name: Name and isotopic mass of the metal ion corresponding to this marker marker_name: Protein name channel_group, channel_group_integer: Biological processes the channel identifies, e.g. specific cell type, signalling, cell death tsne_channel: Logical - use this channel for dimensionality reduction? cluster_channel: Logical - use this channel for clustering?