GenePlexus

GenePlexus predicts associations between genes and user-provided human gene sets using network-based machine learning to support discovery and interpretation of gene functions in biological processes and diseases.

Key Features:

• Network-based machine learning: Uses network-based machine learning to learn patterns of functional relatedness among genes.
• User-provided human gene sets: Accepts uploaded human gene lists as the input seed set for analysis.
• Multiple human network representations: Supports analysis across various human network representations to capture different relationships among genes.
• Genome-wide prediction: Predicts the association of every gene in the chosen network with the input gene set.
• Interpretability via network visualization: Provides network visualization techniques to explore relationships between input genes and functionally similar genes.
• Model comparisons to known gene sets: Compares results to machine learning models trained on thousands of known process/pathway and disease gene sets.

Scientific Applications:

• Gene discovery: Prioritizes candidate genes associated with a user-provided human gene set for downstream validation.
• Functional interpretation: Elucidates potential functions and pathway or process involvement of genes beyond the original list.
• Disease-related gene analysis: Identifies genes potentially related to disease-associated process/pathway gene sets.

Methodology:

Applies network-based machine learning on various human network representations to predict genome-wide gene associations for an input human gene set, incorporates network visualization for interpretability, and compares predictions to models trained on thousands of known process/pathway and disease gene sets.