pyVIPER

pyVIPER enables rank-based enrichment analysis of single-cell RNA sequencing data for VIPER-based protein activity inference at scale.

Key Features:

• Multiple Enrichment Algorithms: pyVIPER supports multiple enrichment analysis algorithms beyond VIPER alone.
• Data Transformation and Postprocessing: pyVIPER includes data transformation and postprocessing modules for analytical flexibility.
• Gene Regulatory Network Data Structure: pyVIPER implements a novel data structure for gene regulatory network manipulation.
• Python Ecosystem Integration: pyVIPER integrates with AnnData, Scanpy, and several widely adopted machine learning libraries.
• Scalability: pyVIPER achieves orders of magnitude runtime reduction compared to VIPER for large datasets.

Scientific Applications:

• Master Regulator Protein Identification: pyVIPER identifies Master Regulator proteins from gene expression data using the VIPER algorithm.
• Cellular State Determinant Elucidation: pyVIPER supports mechanistic elucidation of cellular state determinants through gene regulatory network analysis.
• Large-Scale Single-Cell Dataset Analysis: pyVIPER enables VIPER-based analysis of single-cell RNA sequencing datasets with up to hundreds of thousands of cells.

Methodology:

pyVIPER performs rank-based enrichment analysis of single-cell RNA sequencing data using the VIPER algorithm and alternative enrichment algorithms, with benchmarking against VIPER for runtime performance on large-scale datasets.