Evolving modular genetic regulatory networks with a recursive, top-down approach

Area of Science:

• Synthetic Biology
• Computational Biology
• Systems Biology

Background:

• Designing genetic regulatory networks (GRNs) for specific cellular functions is a key goal in synthetic biology.
• Identifying minimal GRNs that exhibit desired time-series behaviors remains a significant challenge.

Purpose of the Study:

• To develop a 'top-down' computational approach for evolving minimal genetic regulatory networks (GRNs).
• To demonstrate the recursive bootstrapping of larger, modular GRNs from smaller evolved networks.

Main Methods:

• Utilized differential evolution (DE) to evolve interaction coefficients in dense GRNs.
• Implemented an aggressive pruning strategy to remove excess interactions once target behaviors were identified.
• Incorporated a penalty term to encourage network minimality.

Main Results:

• Successfully rediscovered known small GRNs for toggle switch and oscillatory circuits.
• Evolved complex, modular GRNs by using previously identified GRNs as non-evolvable subnetworks.
• The proposed method, with aggressive pruning and penalty terms, identified minimal or near-minimal GRNs, unlike canonical DE methods.

Conclusions:

• The 'top-down' DE approach with pruning is effective for evolving minimal genetic regulatory networks (GRNs).
• This method facilitates the design of complex, modular biological systems for synthetic biology applications.