Reproduction scripts and processed data for "Targeting Pre-Existing Club-Like Cells in Prostate Cancer Potentiates Androgen Deprivation Therapy"

Overview This dataset contains R scripts and processed single-cell RNA-seq data required to reproduce the computational figures in the manuscript: "Targeting Pre-Existing Club-Like Cells in Prostate Cancer Potentiates Androgen Deprivation Therapy" Published in EMBO Molecular Medicine (2025) Contents R Scripts (5 files) Figure1.R - Analysis of LSCmed cells as surrogate of human MSPC Figure2.R - Transcriptomic heterogeneity and clustering of LSCmed cells Figure3.R - RNA velocity and pseudotime trajectory analysis Figure4.R - SCENIC transcription factor regulon analysis (includes EV4) Figure5.R - Pim1 expression analysis Processed Data Files (6 files) Seurat Objects: GOF_306cells.rds (49 MB) - 306 LSCmed cells with clustering and all signature scores (AR, MSPC, ARPC, NEPC, CRPC subtypes, cell type signatures: Luminal, Basal, Club, Hillock, EMT) seu3001_scenic.rds (62 MB) - Seurat object with SCENIC TF activity scores from 10 aggregated runs SCENIC Results: regulon_geneset.rds (136 KB) - Transcription factor regulon gene sets (union of 10 SCENIC runs) Differential Expression & Count Data: GOFcelltype.markers.csv (257 KB) - Pre-computed differential expression markers between castrated and naive conditions gene_count_3001_4271.txt (66 MB) - Gene count matrix for velocyto cell name mapping 3001.loom (22 MB) - Velocyto loom file for RNA velocity analysis Data Description Experimental Design Model: Ptenpc-/- mouse prostate cancer model Technology: Smart-Seq2 single-cell RNA sequencing Cell Type: LSCmed (luminal stem cell-enriched population) Conditions: Naive (intact) vs Castrated (2 months post-castration) Total Cells: 306 cells (192 castrated + 114 naive) Clusters: 3 major LSCmed subpopulations identified Software Requirements R (>= 4.0) Seurat (>= 4.0) SeuratExtend For RNA velocity: Python with scvelo (conda environment) For pseudotime: monocle, M3Drop For SCENIC: Nextflow + pySCENIC