Comparing computational times for simulations when using PBPK model template and stand-alone implementations of PBPK...

Area of Science:

• Pharmacokinetics
• Computational toxicology
• Systems biology

Background:

• Physiologically based pharmacokinetic (PBPK) models are crucial for predicting chemical behavior in the body.
• A PBPK model template was developed to streamline the implementation of diverse PBPK models.
• Understanding factors affecting PBPK simulation speed is essential for efficient toxicological assessments.

Purpose of the Study:

• To identify key factors influencing the computational time of PBPK model simulations.
• To evaluate the impact of various implementation strategies and model features on simulation performance.
• To provide guidance on optimizing PBPK model design for computational efficiency.

Main Methods:

• Conducted timing experiments using PBPK models for dichloromethane and chloroform.
• Compared template-based and stand-alone PBPK model implementations across different exposure scenarios.
• Assessed the influence of parameter treatment (constant vs. time-varying) and model complexity on computational time.

Main Results:

• Treating body weight as a constant parameter reduced simulation time by up to 30%.
• Decreasing the number of state variables by 36% resulted in a 20-35% reduction in computational time.
• The number of output variables and conditional statement implementation had minimal impact on simulation speed.

Conclusions:

• PBPK model design choices, particularly parameter handling and state variable reduction, significantly impact computational efficiency.
• While template-based PBPK models may require more computation time, their flexibility offers substantial human time savings.
• These findings aid in developing, improving, and applying PBPK models for faster and more efficient simulations.