Related Experiment Videos
Coarse-graining and self-dissimilarity of complex networks
Shalev Itzkovitz1, Reuven Levitt, Nadav Kashtan
1Departments of Molecular Cell Biology and Physics of Complex Systems, Weizmann Institute of Science, Rehovot, Israel.
Physical Review. E, Statistical, Nonlinear, and Soft Matter Physics
|February 9, 2005
Summary
This study introduces a method to simplify complex networks by identifying connectivity patterns as nodes. This approach creates understandable, multi-level maps for both electronic circuits and biological systems.
Area of Science:
- Network science
- Systems biology
- Electrical engineering
Background:
- Complex engineered and biological networks often lack clear, hierarchical organization.
- Understanding these networks requires methods to reduce their complexity without losing essential information.
Purpose of the Study:
- To develop and present algorithms for coarse-graining complex networks.
- To demonstrate the application of this coarse-graining approach to electronic circuits and biological networks.
- To investigate the hierarchical structure and self-dissimilarity of networks.
Main Methods:
- Defining coarse-graining units as connectivity patterns within networks.
- Developing algorithms to detect these connectivity patterns.
- Applying the coarse-graining algorithms to reverse-engineer electronic circuits and analyze mammalian protein signal-transduction networks.
Main Results:
- Successfully created multi-level, understandable maps of electronic circuits, from transistor-level to module-level representations.
- Identified a simplified coarse-grained network for mammalian protein signaling, revealing three main signaling channels resembling multi-layered perceptrons.
- Demonstrated that both electronic and biological networks exhibit 'self-dissimilarity,' with distinct network motifs at different levels of abstraction.
Conclusions:
- The developed coarse-graining approach effectively simplifies complex engineered and biological networks.
- This method provides a powerful tool for creating hierarchical, understandable representations of intricate systems.
- The findings suggest a universal principle of network organization applicable to diverse natural and designed systems.