A partial-propensity variant of the composition-rejection stochastic simulation algorithm for chemical reaction...

Area of Science:

• Computational chemistry
• Biochemical systems analysis
• Stochastic modeling

Background:

• Stochastic simulation algorithms (SSA) are crucial for modeling well-stirred chemical reaction systems.
• Existing SSA methods can face computational challenges with complex or large reaction networks.
• Partial-propensity methods and composition-rejection SSA offer specific advantages but have limitations.

Purpose of the Study:

• To develop a novel, exact stochastic simulation algorithm for coupled chemical reactions.
• To improve the computational efficiency and scalability of SSA for diverse reaction network structures.
• To combine the benefits of partial-propensity approaches and composition-rejection sampling.

Main Methods:

• Implementation of the partial-propensity stochastic simulation algorithm with composition-rejection sampling (PSSA-CR).
• Adaptation of existing composition-rejection SSA framework with partial-propensity calculations.
• Analysis of computational cost scaling for weakly and strongly coupled reaction networks.

Main Results:

• The PSSA-CR algorithm provides an exact formulation for well-stirred chemical reaction systems.
• Computational cost is bounded by a constant for weakly coupled networks.
• Computational cost scales at most linearly with the number of species for strongly coupled networks.

Conclusions:

• PSSA-CR effectively merges the advantages of partial-propensity methods and composition-rejection SSA.
• The algorithm demonstrates favorable computational cost scaling across all classes of reaction networks.
• This advancement offers a more efficient tool for stochastic simulation in chemical kinetics and systems biology.