Principles of Computation by Competitive Protein Dimerization Networks.

Area of Science:

• Biochemistry
• Systems Biology
• Computational Biology

Background:

• Biological signaling pathways frequently utilize proteins that competitively dimerize.
• These dimerization networks function as biochemical computers, translating monomer concentrations (inputs) into dimer concentrations (outputs).
• The computational capabilities, or
• expressivity
• of these networks are not well understood.

Purpose of the Study:

• To investigate the computational range and versatility of protein dimerization networks.
• To determine how network size and connectivity influence their input-output computations.
• To explore the potential of dimerization networks for signal processing.

Main Methods:

• A systematic computational approach was employed.
• Analysis focused on network size, connectivity, and protein expression levels.
• Simulations assessed the range of biochemical computations performed.

Main Results:

• Small dimerization networks (3-6 monomers) exhibit significant expressivity, performing diverse multi-input computations.
• These networks demonstrate versatility, altering computations based on protein expression levels (e.g., in different cell types).
• Larger networks (≥8 proteins) with random affinities can perform approximately 90% of potential one-input computations by adjusting monomer expression.

Conclusions:

• Competitive protein dimerization is a powerful and versatile mechanism for biochemical computation.
• Dimerization networks offer a robust architecture for multi-input, cell-type-specific signal processing.
• The study highlights the significant computational potential inherent in simple dimerization processes.