Processed data for Wei et al. "High-confidence calling of normal epithelial cells allows identification of a novel stem-like cell state in the colorectal cancer microenvironment" the scanpy object contains raw (cellbender) and normalized counts, two UMAPs for all cells and epithelial cells only, respectively, and the following meta-data columns: sample: sample ID sample_origin: tumour or normal patient: patient ID MS_status: MSI or MSS n_counts: # of UMIs per cell n_genes: # of gnes per cells S_score: score for S phase G2M_score: score of G2/M phase cell_type_coarse: low-resolution cell type annotation (epithelial vs. immune vs. stromal) cell_type_fine: high-resolution cell type annotation (by label transfer from Uhlitz et al.) CCISM_call: cell assignment by CCISM CCISM_score: score from CCISM numbat_call: assignment by numbat numbat_score: score from numbat inferCNV_call: assignment from inferCNV iCMS_call: assignment using iCMS signatures consensus_call: numbat/CCISM consensus identity cell_type_epi: epithelial cell type (combining TC1-4 from Uhlitz et al., and labeling cells defined here as genomically cancer as Tumour) cytotrace_pseudotime: pseudotime from cytotrace louvain: louvain clusters for epithelial cells list of processed data used to create this object folder cellranger contains raw feature barcode matrices for single-cell RNAseq data of 10 patients folder cellbender contains filtered feature barcode matrices after background removal folder WGS contains vcf.gz results from somatic variant calling with mutect2 folder cellSNP contains `cellsnp-lite` output: counts of total and alternative alleles observed for somatic variants in scRNAseq data; subfolder downsampled contains results after downsampling bam files to 5, 10, 20 or 50% of reads folder numbat contains final clone assignments by Numbat