Abstract 2501: pyVIPER: Enabling scalable and rapid insights from single-cell RNASeq data in cancer research

Abstract Single-cell sequencing has transformed cancer research by enabling the study of tumor heterogeneity and the complex interactions within the tumor microenvironment at an unprecedented resolution. However, the low signal-to-noise ratio in single-cell RNA sequencing (scRNASeq) poses challenges for quantitatively deciphering key cancer processes. We have demonstrated that gene regulatory network (GRN) analysis can address this by identifying Master Regulator (MR) proteins, using the VIPER algorithm, to pinpoint critical drivers of cancer and its microenvironment. As scRNASeq datasets from patient-derived tumors grow in size and complexity, there is a pressing need for scalable tools capable of analyzing datasets with hundreds of thousands of cells. To meet this need, we introduce pyVIPER, a highly efficient and scalable Python toolkit designed to assess protein activity in large-scale scRNASeq datasets. pyVIPER facilitates tumor and microenvironment analysis by supporting multiple enrichment algorithms, advanced data transformations, and a novel data structure for GRN manipulation. With seamless integration into publicly available state-of-art scRNASeq tools, as well as machine learning libraries, pyVIPER provides the ideal platform for analyzing tumor-derived samples. Benchmarking shows that pyVIPER reduces runtime from hours to minutes for large datasets, compared to previous tools, enabling a fast and comprehensive analysis of cancer and tumor microenvironment dynamics at single-cell resolution. This advancement makes VIPER-based analysis accessible for virtually any large-scale single-cell dataset from patient tumors, offering new insights into cancer biology and therapeutic targets. Citation Format: Alexander L. Wang, Zizhao Lin, Luca Zanella, Lukas Vlahos, Miquel Anglada Girotto, Aziz Ur Rehman Zafar, Heeju Noh, Andrea Califano, Alessandro Vasciaveo. pyVIPER: Enabling scalable and rapid insights from single-cell RNASeq data in cancer research [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2501.