Comparing signaling networks between normal and transformed hepatocytes using discrete logical models

Area of Science:

• Systems biology
• Computational biology
• Molecular biology

Background:

• Large-scale gene and protein networks offer insights into biological systems but lack context specificity.
• Traditional cell signaling analysis is limited and cannot easily incorporate network-level data.
• Predicting cellular responses to stimuli requires integrating network topology with functional data.

Purpose of the Study:

• To develop a hybrid approach combining network analysis and functional experimentation for cell signaling.
• To create context-specific mathematical models of immediate-early signaling in liver cells.
• To differentiate between normal and diseased liver cell signaling pathways.

Main Methods:

• Constructed Boolean logic models from literature-based prior knowledge networks.
• Trained models against biochemical data from primary human hepatocytes and hepatocellular carcinoma cell lines.
• Utilized diverse stimuli including cytokines and small-molecule kinase inhibitors.

Main Results:

• Generated distinct families of Boolean models for different liver cell types.
• Successfully trained models against experimental biochemical data.
• Observed topological clustering of models into normal and diseased sets, indicating distinct signaling states.

Conclusions:

• The hybrid approach effectively models context-specific cell signaling.
• Boolean logic models can capture differences between normal and diseased liver cell signaling.
• This method provides a framework for analyzing complex biological networks and predicting cellular responses.