What problem does it solve? Single-cell RNA-seq analysis involves many error-prone steps—QC, normalization, clustering, integration, annotation, and reporting—and FGCZ analysts must additionally follow strict institutional conventions for plots, storage paths, and R Markdown reports. This Skill encodes those workflows and standards so analyses are reproducible and consistent. ## Core Features & Use Cases - Standard scRNA-seq pipeline: Load data from h5ad, 10x CellRanger, or qs2 objects, then run QC, normalization, PCA, UMAP, clustering, and differential expression with an automated R script. - Multi-sample integration: Integrate datasets with Harmony (recommended), CCA, RPCA, FastMNN, or scVI, including parameter tuning and troubleshooting guidance. - Advanced analyses: Compositional analysis with sccomp, pseudo-bulk PCA, MSE distance analysis, Augur cell-type prioritization, and CytoTRACE 2 potency scoring. - FGCZ-standard reporting: Generate R Markdown HTML reports with the ezRun template, polychrome cluster colors, coord_fixed UMAPs, interactive DEG tables, and 300 DPI figures. - Use Case: A bioinformatician receives CellRanger outputs for project p12345, runs the QC pipeline script, integrates samples with Harmony, annotates cell types, and delivers an FGCZ-branded HTML report saved to gStore via g-req. ## Quick Start Ask the agent to load your h5ad or CellRanger data and run the Seurat QC and clustering pipeline following FGCZ conventions, then generate an R Markdown report.
