Designing synthetic networks in silico: a generalised evolutionary algorithm approach.

Area of Science:

• Systems and Synthetic Biology
• Computational Biology
• Evolutionary Algorithms

Background:

• Biological networks evolve in response to environmental signals.
• Key aims include elucidating, understanding, and reconstructing biological network motifs.
• Previous research focused on optimizing network structures and reaction rates for specific dynamic behaviors.

Purpose of the Study:

• To present a generalized in silico evolutionary algorithm for biological network design.
• To simultaneously optimize network structures and reaction rates (genotypes) for multiple objectives (phenotypes).

Main Methods:

• Translating biological network descriptions into systems of ordinary differential equations (ODEs).
• Numerically solving ODEs to analyze network functionality.
• Benchmarking algorithm performance using the Repressilator model and parameter optimization.

Main Results:

• Successfully recapitulated concentration time-series data and optimized oscillatory dynamics.
• Identified novel designs for robust synthetic oscillators.
• Performed multi-objective optimization to balance different system properties in oscillators and feed-forward loops.

Conclusions:

• An evolutionary algorithm was developed and tested for designing biological networks with desired outputs.
• This algorithm expands the accessible design space beyond previous limitations.
• Enables synthetic biologists to construct systems with a wider range of complex responses.