scGraphVerse

scGraphVerse infers, compares, and visualizes gene networks from single-cell RNA sequencing data.

Key Features:

• Multi-method network inference: Integrates GENIE3, GRNBoost2, ZILGM, PCzinb, and JRF for robust gene regulatory network inference.
• Consensus building: Supports consensus network construction across methods or datasets.
• Network evaluation: Provides tools for evaluating regulatory structure and community similarity.
• Python dependency management: GRNBoost2 requires Python package 'arboreto' which can be installed using init_py(install_missing = TRUE).
• Adapted functions: Includes adapted functions from ZILGM (Park et al., 2021), JRF (Petralia et al., 2015), and learn2count (Nguyen et al. 2023) packages with proper attribution under GPL-2 license.

Scientific Applications:

• Single-cell gene network analysis: Infers gene regulatory networks from single-cell RNA sequencing data.
• Cross-method validation: Compares network inference results across multiple algorithms to improve robustness.
• Cross-dataset integration: Builds consensus networks by integrating results from different datasets.
• Regulatory structure analysis: Evaluates the architecture and organization of gene regulatory networks.
• Community similarity assessment: Quantifies similarity between network communities or modules.

Methodology:

scGraphVerse integrates multiple network inference methods including GENIE3, GRNBoost2, ZILGM, PCzinb, and JRF, and supports consensus building across methods or datasets for robust gene network inference from single-cell RNA sequencing data.