PTNet

PTNet predicts protein levels by integrating miRNA-mRNA interaction networks with mRNA and miRNA expression data to model miRNA-mediated post-transcriptional regulation.

Key Features:

• Graph-Based Learning Model: PTNet models interactions among miRNAs, mRNAs, and proteins using a graph-based learning approach to represent post-transcriptional regulation.
• Simulation of miRNA-Mediated Regulation: PTNet simulates miRNA-controlled regulatory networks in silico using the miRNA-mRNA interaction network together with expression profiles.
• Predictive Capability: PTNet predicts protein abundance that correlates more closely with proteomic data than mRNA expression alone, supporting analyses such as disease prognosis.

Scientific Applications:

• Disease Pathogenesis Studies: PTNet aids study of molecular mechanisms in diseases such as cancer by providing protein-level predictions from transcriptomic and miRNA data.

Methodology:

PTNet applies a graph-based learning model that integrates the miRNA-mRNA interaction network with mRNA and miRNA expression profiles to simulate miRNA-mediated gene regulation in silico without relying on large-scale proteomics data.