A new version of MULTI (ELS) for extended nonlinear least squares

Area of Science:

• Pharmacokinetics
• Computational Biology
• Software Development

Background:

• Population pharmacokinetics (PPK) is crucial for understanding drug behavior in diverse patient groups.
• Microcomputers are increasingly used for complex pharmacokinetic analyses.
• Existing software may lack flexibility in parameter constraints and algorithmic approaches.

Purpose of the Study:

• To introduce a new version of the MULTI (Extended Least Squares) software, rewritten in BASIC for microcomputer accessibility.
• To enhance population pharmacokinetic modeling capabilities with advanced algorithms and parameter transformations.

Main Methods:

• The MULTI (ELS) software utilizes four extended nonlinear least squares algorithms: steepest descent, quasi-Newton (DFP and BFGS), and simplex methods.
• It incorporates four parameter transformation options to constrain population parameters, including no constraints, squared transformation, sinusoidal transformation, and logistic transformation.
• The software is implemented in BASIC, ensuring compatibility with microcomputers.

Main Results:

• The new MULTI (ELS) version provides a user-friendly interface for population pharmacokinetic analysis on microcomputers.
• The inclusion of multiple nonlinear least squares algorithms and flexible parameter transformations allows for robust and adaptable pharmacokinetic model fitting.
• The software maintains the established definitions of population models and modifications for BASIC compilation from its previous version.

Conclusions:

• The updated MULTI (ELS) software offers a powerful and accessible tool for population pharmacokinetic modeling on microcomputers.
• Its advanced algorithms and parameter constraint options facilitate more accurate and reliable drug disposition analysis.
• This development supports broader application of population pharmacokinetics in research and clinical settings.