Exploring biological network structure with clustered random networks.

Area of Science:

• Network science
• Computational biology
• Systems biology

Background:

• Complex biological systems are often modeled as networks.
• Network properties like node degree and clustering influence system dynamics.
• Disentangling interdependent network effects is challenging.

Purpose of the Study:

• To develop a new algorithm for generating random graphs with specified degree sequence and clustering.
• To provide a tool for creating accurate null models in bioinformatics research.
• To enable systematic study of network structure's impact on function and dynamics.

Main Methods:

• Developed and implemented a Markov chain simulation algorithm.
• The algorithm, ClustRNet (Clustered Random Networks), generates simple, connected random graphs.
• Optimized graph generation for local or global clustering measures.

Main Results:

• ClustRNet successfully generates random graphs with desired degree sequence and clustering.
• The algorithm outperforms existing methods in producing specified network characteristics.
• Generated networks serve as effective random controls for empirical network analysis.

Conclusions:

• ClustRNet generates ensembles of graphs with specified edge structure and clustering.
• These graphs facilitate the study of connectivity and redundancy impacts on network function.
• This is a key step in understanding biological systems' structure-function relationships.