Skip to main content Skip to navigation

Project 5 - Deconvolution in Non-linear Ordinary Differential Equations for QSP


ESR - Magnus Tragardh

Aims

The aim of the project is to investigate, develop and apply methods for estimating the input to a dynamical system, based on sparse and noisy measurements of certain state variables. This is of interest in a variety of contexts in drug development. A typical example would be to estimate the absorption rate and bioavaliability of a drug after oral administration, based on measurements of the drug concentration in blood plasma. Traditional methods for input estimation in pharmacokinetics tend to focus on linear models. However, many models on interest contain nonlinearities, and it is therefore desirable to develop methods that can handle that case too.

Supervisor: Peter Gennemark, AstraZeneca

Co-Supervisors: Johan Gabrielsson (SUAS), Mats Jisrtrand (FCC), Mike Chappell (University of Warwick), Neil Evans (University of Warwick)

 Project Objectives
  • The development and improvement of methodologies for identifying unknown mathematical relationships in a (nonlinear) pharmacological system where only certain state variables are observed and where data are sparse and noisy.
  • Develop improved deconvolution (identification) techniques for quanitative and predictive models of covering the following stages of drug development: drug administration, drug exposure, target engagement and pharmacological response.
  • An improved understanding of system properties, thus increasing confidence in the selection between prospective compounds and providing more robust design tools for drug discovery.

Publications

  • Trägårdh M, Chappell M. J., Palm J. D., Evans N. D., Janzén D. L. I., Gennemark P. (2017). ‘Input estimation for Extended Release Formulations Exemplified with Exenatide’. Frontiers in Bioengineering and Biotechnology: Computational Physiology and Medicine. doi: 10.3389/fbioe.2017.00024
  • Trägårdh M., Chappell, M. J., Ahnmark A., Lindén, D., Evans N. D., Gennemark P. (2016). Input estimation for drug discovery using optimal control and Markov chain Monte Carlo approaches. Journal of Pharmacokinetics and Pharmacodynamics. doi:10.1007/s10928-016-9467-z Supporting code: idiom-1.0.0.zip This software is downloadable under an academic use licence. For any use other than academic research, please use the contact details in the licence. The full licence is available here.

Conference Papers

  • 6th-9th September 2016: ‘Input estimation for drug discovery using optimal control and Markov Chain Monte Carlo approaches’. QSP UK 1st Exchange Workshop, University of Surrey, Guildford.

Poster Presentations

  • Trägårdh M, Chappell M, Evans N, Gennemark P: Input estimation for drug discovery using optimal control and Markov Chain Monte Carlo approaches. Presented at PKUK 2015.
  • Trägårdh M, Chappell M, Evans N, Jirstrand M, Gabrielsson J, Gennemark P: Input estimation in nonlinear dynamical systems for model-based drug discovery using optimal control techniques. Presented at the AstraZeneca Modelling & Simulation Symposium 2014 and PKUK 2014.
  • Andersson R, Janzén D, Trägårdh M, Chappell M, Evans N, Gabrielsson J, Gennemark P, Jirstrand M, Peletier L, Yates J: Dose-Response-Time Modelling, Input Estimation and Structural Identifiability as tools for Systems Pharmacology. Presented at the the 7th Noordwijkerhout Symposium on Pharmacokinetics, Pharmacodynamics and Systems Pharmacology 2014 and at the Chalmers Life Science Area of Advance Meeting 2014.
  • Trägårdh M, Chappell M, Evans N, Jirstrand M, Gabrielsson J, Gennemark P: Input Estimation in Nonlinear Ordinary Differential Equations for Quantitative and Systems Pharmacology. Presented at PKUK 2013 and at AstraZeneca Modelling & Simulation 2013.

Other Presentations

  • 6th November 2015: One of the speakers on ‘Emergent M&S-related methods’. AstraZeneca, Mölndal, Sweden.
  • 20th May 2014: Input estimation in nonlinear dynamical systems for Quantitative and Systems Pharmacology. AstraZeneca, Mölndal, Sweden.

Conferences Attended

  • 18th-20th November 2015: The PKUK (Pharmacokinetics UK) meeeting in Chester, .
  • 5th–7th November 2014:The PKUK (Pharmacokinetics UK) meeting in Bath.
  • 5th May 2014: Chalmers Life Science Engineering Area of Advance Meeting at Chalmers University of Technology, Gothenburg, Sweden.
  • 23rd–25th April 2014: The 7th Noordwijkerhout Symposium on Pharmacokinetics, Pharmacodynamics and Systems Pharmacology.
  • 1st–2nd November 2013: The PKUK (Pharmacokinetics UK) meeting in Harrogate.