We profile the transcriptomes of ~30,000 mouse single cells to deconvolve the hepatic mesenchyme in healthy and fibrotic liver at high resolution. We reveal spatial zonation of hepatic stellate cells across the liver lobule, designated portal vein-associated HSC and central vein-associated HSC, and uncover an equivalent functional zonation in a mouse model of centrilobular fibrosis. Our work illustrates the power of single-cell transcriptomics to resolve key collagen-producing cells driving liver fibrosis with high precision. We provide the contents of these data as Seurat R objects.uninjured.rds: Seurat object in RDS format containing the raw, normalised, and scaled expression values for 12533 mouse single cell transcriptomes from healthy liver, corresponding to Figure 1. injured.rds: Seurat object in RDS format containing the raw, normalised, and scaled expression values for 23291 mouse single cell transcriptomes from healthy liver, corresponding to Figure 3. acute.rds: Seurat object in RDS format containing the raw, normalised, and scaled expression values for 7260 mouse single cell transcriptomes from healthy liver, corresponding to Figure 5. ss2.rds: Seurat object in RDS format containing the raw, normalised, and scaled expression values for 905 mouse single cell transcriptomes from healthy liver, corresponding to Figure S3. The Seurat objects contain metadata fields as follows: - various quality control metrics such as nGene, nUMI, percent.mito - enrichment: isolation protocol used, either PF (digestion protocol 1) or HSC (digestion protocol 2) - dataset: dataset of origin, i.e. HSC_Uninjured, PF_CCl4_Peak - condition: condition of liver of origin, i.e. Uninjured, Peak CCl4 - lineage: annotated mesenchymal lineage, either FB, HSC, or VSMC - hsc_zonation: annotated HSC zonation, either Portal or Central - pseudotime_central and pseudotime_portal (acute only): annotated pseudotemporal values