Matthew J. Simpson

This study develops a likelihood-based workflow to predict extinction risk in bistable populations using noisy count data, showing that data quality and quantity significantly affect prediction accuracy.

This study develops a mathematical model to show how curvature-dependent tissue growth arises from cell crowding and mechanical interactions, predicting pore closure scaling with size and shape.

The paper develops a neural posterior estimation framework for parameter inference in spatiotemporal stochastic biological models, demonstrating its effectiveness across multiple cell migration models.

This paper presents a computational model to simulate tissue growth by integrating geometry, mechanics, and stochastic processes, demonstrating its application to 3D-printed scaffolds and bone formation in cortical pores.

The paper introduces parameter-wise prediction for stochastic random walk models in life sciences, using PDE surrogates and likelihood-based methods to analyze sensitivity and parameter estimation.

This paper reviews parameter inference, identifiability analysis, and model prediction for stochastic agent-based models of biological populations, with a focus on numerical optimization and automatic differentiation, and provides open-source Julia code.

The paper presents a computational approach for reducing non-identifiable mathematical models using likelihood reparameterisation to enable efficient predictions.

This paper applies the Profile-Wise Analysis workflow to non-identifiable mathematical models in population growth to improve prediction and parameter estimation.

The paper compares Gaussian and binomial measurement error models for PDE-based biological population dynamics, finding the binomial model provides biologically plausible predictions.

The paper introduces a workflow called Profile-Wise Analysis for identifiability analysis, parameter estimation, and prediction using mechanistic mathematical models, demonstrating its application with ODE models.

The paper develops a computationally efficient framework for diagnosing, understanding, and predicting biphasic population growth using profile likelihood analysis.

This paper reviews optimal control theory in systems biology, implements the forward-backward sweep method, and accelerates its convergence using existing techniques.

This paper analyzes identifiability in stochastic differential equation models using case studies and synthetic data, showing they can extract more parameter information than deterministic models.

The paper develops a theoretical framework for understanding bacterial persistence as a hedging strategy by integrating optimal control theory from finance and mathematical biology.

This paper applies optimal control theory to design combination therapies for acute myeloid leukaemia by analyzing control dynamics and pay-off functions.

This paper develops optimal chemotherapy strategies for AML by incorporating immune response into a stem cell model and applying optimal control theory.

The paper demonstrates that process noise can distinguish between indistinguishable stochastic population dynamics models using Bayesian methods and summary statistics.

The paper develops a free boundary model to describe epithelial dynamics by deriving a continuum-limit PDE that accurately predicts cell-based model behavior with a mass-conserving boundary condition.