Poster #10: Tanush Parasam Harish MS

scPipelinePy: A Modular Open-Source Python Framework for Reproducible Single-Cell RNA-seq Analysis

Tanush Parasam Harish

Single-cell RNA sequencing (scRNA-seq) has become a cornerstone technology for characterizing cellular heterogeneity, but existing analysis workflows often require extensive manual configuration, lack standardized documentation, and present reproducibility challenges across research groups. To address these barriers, we developed scPipelinePy, a modular, open-source Python framework for end-to-end scRNA-seq analysis. Methods: scPipelinePy implements a complete analytical workflow spanning quality control (gene/UMI count filtering, mitochondrial content thresholding, and doublet detection), normalization, dimensionality reduction, and graph-based clustering. The package comprises over 10 documented functions built on the Scanpy and AnnData ecosystem, supporting multiple input formats including 10X Genomics, H5, and CSV files. Users can customize QC thresholds and clustering parameters to accommodate diverse experimental designs, and the pipeline automatically generates UMAP visualizations and cluster-level summary statistics without requiring manual plotting code. Results: We validated scPipelinePy on a public peripheral blood mononuclear cell (PBMC) benchmark dataset comprising approximately 70,000 cells. Leiden clustering combined with marker gene expression analysis achieved greater than 95% concordance with previously published cell type annotations, demonstrating that the pipeline reliably recovers biologically meaningful cell populations using standardized, minimal-configuration parameters. To support adoption and reproducibility, we provide comprehensive documentation, including Jupyter notebook tutorials that walk users through the full analytical workflow, Docker containerization for consistent deployment across computing environments, and interactive HTML reports summarizing QC metrics and dimensionality reduction results for each dataset processed. Conclusion: scPipelinePy lowers the barrier to rigorous, reproducible scRNA-seq analysis by consolidating best-practice methods into a single, well-documented, and containerized framework. Its modular design allows researchers to adapt individual pipeline components to dataset-specific needs while maintaining methodological consistency, making it a practical tool for both new and experienced computational biologists working with single-cell data.