2026-10-08T12:01:40Z https://api.figshare.com/v2/oai

oai:figshare.com:article/34030473 2026-09-30T07:13:55Z category_1 category_4 category_915 category_7 category_873 category_734 category_931 category_811 category_135 portal_63 item_type_6 month_year_09_2026

ChemODE: A Physics-Informed Neural Surrogate for Robust Pharmacokinetics and Biochemical Dynamics Wit Kulvutiroj (25142313) Biophysics Biochemistry Physical Sciences not elsewhere classified Medicine Chemical Sciences not elsewhere classified Biological Sciences not elsewhere classified Information Systems not elsewhere classified Mathematical Sciences not elsewhere classified Computational Biology rigorous uncertainty quantification resolve magnitude bias rational architectural constraints polynomial basis functions nonlinear saturation limits neural residual paradigm multicompartment scale separations menten kinetics ). governing kinetic laws fundamental biological benchmarks fly feature normalization extrapolate physical saturation demonstrate clinical applicability constrained rational layers box framework tailored interpretable mechanism discovery driven mechanism discovery informed neural surrogate based optimization scheme max </ sub high precision (< dominated biochemical settings compartment pharmacokinetic model automated discovery v </ informed gray based minimization >< sub systems biology sparse sampling seed ensembles robust pharmacokinetics fundamentally fail experimental domains elimination rates critical challenge chemical engineering biochemical dynamics The automated discovery of governing kinetic laws from observational data is a critical challenge in systems biology, pharmacology, and chemical engineering. In these experimental domains, data-driven mechanism discovery is complicated by sparse sampling, measurement noise (e.g., blood assays), and the prevalence of nonlinear saturation limits (Michaelis-Menten kinetics). Existing symbolic regression methods, such as SINDy, rely on polynomial basis functions that fundamentally fail to extrapolate physical saturation, and their reliance on numerical differentiation degrades catastrophically in high-noise regimes (>5%). In this work, we introduce ChemODE, a physics-informed gray-box framework tailored for biochemical and pharmacokinetic (PK) networks. ChemODE integrates (1) Constrained Rational Layers to autonomously enforce physical saturation laws, (2) On-the-fly Feature Normalization to resolve magnitude bias in multicompartment scale separations, and (3) a noise-robust integral-based optimization scheme. We validate ChemODE on fundamental biological benchmarks, including the nonlinear Brusselator limit cycle and a synthetic 3-compartment pharmacokinetic model. To demonstrate clinical applicability, ChemODE successfully extracts physiological absorption and elimination rates from real-world human theophylline pharmacokinetic assays. Furthermore, rigorous uncertainty quantification (10-seed ensembles) proves that ChemODE recovers enzyme kinetics with high precision (<i>V</i>_max error ≈ 1%), significantly outperforming unconstrained baselines. This work establishes that integral-based minimization, combined with rational architectural constraints and a neural residual paradigm, provides a robust surrogate modeling path for interpretable mechanism discovery in noise-dominated biochemical settings. 2026-09-30T00:00:00Z Text Journal contribution 10.1021/acsomega.6c04487.s001 https://figshare.com/articles/journal_contribution/ChemODE_A_Physics-Informed_Neural_Surrogate_for_Robust_Pharmacokinetics_and_Biochemical_Dynamics/34030473 CC BY-NC 4.0