Simulation-based Methods for Optimal Sampling Design in Systems Biology
Tuan Minh Ha, Binh Thanh Nguyen, Lam Si Tung Ho
TL;DR
This paper tackles the problem of optimal sampling design for parameter estimation in dynamical systems used in systems biology, addressing the limitation that classical $FIM$-based criteria require an initial parameter guess. It introduces two simulation-based approaches—E-optimal-ranking ($EOR$) and an attention-based Long Short-Term Memory ($At$-LSTM)—that do not depend on a prior parameter estimate. The methods are demonstrated on Lotka–Volterra and a three-compartment pharmacokinetic model, where they outperform random sampling and the standard $E$-optimal design, with $EOR$ and $At$-LSTM showing complementary strengths. These findings suggest robust, uncertainty-tolerant sampling strategies and open avenues for replacing or complementing classical OED with ensemble or alternative ML-driven designs.
Abstract
In many areas of systems biology, including virology, pharmacokinetics, and population biology, dynamical systems are commonly used to describe biological processes. These systems can be characterized by estimating their parameters from sampled data. The key problem is how to optimally select sampling points to achieve accurate parameter estimation. Classical approaches often rely on Fisher information matrix-based criteria such as A-, D-, and E-optimality, which require an initial parameter estimate and may yield suboptimal results when the estimate is inaccurate. This study proposes two simulation-based methods for optimal sampling design that do not depend on initial parameter estimates. The first method, E-optimal-ranking (EOR), employs the E-optimal criterion, while the second utilizes a Long Short-Term Memory (LSTM) neural network. Simulation studies based on the Lotka-Volterra and three-compartment models demonstrate that the proposed methods outperform both random selection and classical E-optimal design.