Methods of information theory and algorithmic complexity for network biology

Area of Science:

• Network biology
• Information theory
• Computational biology

Background:

• Information theory provides tools to quantify complex systems.
• Network biology analyzes biological systems as networks.
• Bridging these fields offers new insights into biological data.

Purpose of the Study:

• To survey and introduce concepts at the intersection of information theory and network biology.
• To demonstrate how information theory quantifies biological data.
• To define and validate robust graph complexity measures.

Main Methods:

• Applying Shannon entropy, compressibility, and algorithmic complexity to biological data.
• Performing exact theoretical calculations and numerical approximations for graph properties.
• Simulating gene expression data to recover topological properties.

Main Results:

• Information theory measures quantify local and global aspects of biological networks.
• Demonstrated recovery of topological properties from simulated gene expression data.
• Introduced formal graph complexity definitions and validated them using Kolmogorov complexity.

Conclusions:

• Information theory offers powerful tools for analyzing network biology.
• The proposed graph complexity measures are robust and applicable to biological networks.
• This interdisciplinary approach enhances understanding of biological system complexity.