netprioR

netprioR prioritizes genes using semi-supervised learning by integrating network data, phenotypic information, and prior true positive (TP) and true negative (TN) gene labels to identify biologically relevant genes within complex biological networks.

Key Features:

• Integration of Network Data: Incorporates network topology and interactions between genes and other molecular entities to provide functional context for prioritization.
• Phenotypic Information Utilization: Integrates phenotypic data to link genetic signals with observable traits or disease states.
• Incorporation of Prior Knowledge: Uses prior TP and TN gene labels sourced from literature or expert annotations to inform and ground predictions.
• Semi-Supervised Learning Approach: Applies a semi-supervised framework that balances labeled and unlabeled data to improve robustness and generalizability of rankings.

Scientific Applications:

• Disease Gene Identification: Prioritizes genes associated with specific phenotypes to support identification of candidate disease genes.
• Functional Genomics Studies: Enables exploration of gene functions within biological pathways and interaction networks.
• Drug Target Discovery: Identifies key genes involved in disease processes that may serve as potential drug targets.

Methodology:

Implements a semi-supervised learning paradigm that combines network topology, phenotypic correlations, and prior TP/TN labels across multiple data sources, leveraging both labeled and unlabeled data for gene ranking.