PRYNT

PRYNT prioritizes disease candidate proteins from proteomic datasets by leveraging contextualized protein-protein interaction (PPI) networks to infer mechanistic links between biofluid changes (e.g., urine) and tissue pathology (e.g., kidney).

Key Features:

• Algorithmic approach: Combines two closeness-based algorithms—shortest-path and random walk—to score and prioritize proteins within a network.
• PPI network construction: Builds a contextualized protein-protein interaction network using clique consolidation of interactions from the STRING database.
• Contextualization: Leverages the context-specific PPI network to identify proteins indirectly associated with disease mechanisms from biofluid proteomes.
• Performance evaluation: Validated on four urinary proteome datasets, demonstrating improved precision and specificity for prioritizing kidney disease candidates and complementarity to upstream regulator analysis tools such as Ingenuity Pathway Analysis.

Scientific Applications:

• Nephrology biomarker discovery: Prioritizes urinary proteome candidates for identification of kidney disease biomarkers.
• Mechanistic inference from biofluids: Infers key proteins indirectly linked to tissue pathology from biofluid proteomic changes to support mechanistic hypotheses.
• Cross-condition applicability: Applicable to other biofluids, molecular traits, and diseases beyond nephrology.

Methodology:

Constructs a contextualized PPI network from STRING via clique consolidation and applies shortest-path and random-walk closeness algorithms to prioritize proteins.