Systems Biology: Identifiability Analysis and Parameter Identification via Systems-Biology-Informed Neural Networks

Mitchell Daneker; Zhen Zhang; George Em Karniadakis; Lu Lu

doi:10.1007/978-1-0716-3008-2_4

Systems Biology: Identifiability Analysis and Parameter Identification via Systems-Biology-Informed Neural Networks

Methods Mol Biol. 2023:2634:87-105. doi: 10.1007/978-1-0716-3008-2_4.

Authors

Mitchell Daneker¹, Zhen Zhang², George Em Karniadakis^{2

3}, Lu Lu⁴

Affiliations

¹ Department of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, PA, USA.
² Division of Applied Mathematics, Brown University, Providence, RI, USA.
³ School of Engineering, Brown University, Providence, RI, USA.
⁴ Department of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, PA, USA. lulu1@seas.upenn.edu.

PMID: 37074575
DOI: 10.1007/978-1-0716-3008-2_4

Abstract

The dynamics of systems biological processes are usually modeled by a system of ordinary differential equations (ODEs) with many unknown parameters that need to be inferred from noisy and sparse measurements. Here, we introduce systems-biology-informed neural networks for parameter estimation by incorporating the system of ODEs into the neural networks. To complete the workflow of system identification, we also describe structural and practical identifiability analysis to analyze the identifiability of parameters. We use the ultradian endocrine model for glucose-insulin interaction as the example to demonstrate all these methods and their implementation.

Keywords: Parameter estimation; Physics-informed neural networks; Practical identifiability; Structural identifiability; Systems biology.

MeSH terms

Models, Biological*
Neural Networks, Computer
Systems Biology* / methods