Dynamic modelling with R: Tools and models for understanding biological systems

Date

Jan 12, 2012

Time

2:30 PM - 3:30 PM

Speaker

Thomas Petzoldt

Affiliation

Technische Universität Dresden, Institut für Hydrobiologie, 01062 Dresden

Series

IMB - Seminar

Language

en

Main Topic

Mathematik

Other Topics

Biologie, Mathematik, Informatik, Medizin

Description

In recent years, the R software has become a well known statistics tool for many scientists, especially at the PhD and the postdoc level. However, R is not only a statistics and graphics system, it is a multi-purpose high-level programming language for scientific computing in general. An increasing collection of packages explicitly developed for dynamic modelling (e.g. deSolve, simecol, ReacTran, FME, cf. Soetaert et al 2010) and a growing number of textbooks teaching systems understanding by using R are indicators for this trend. The presentation will show examples demonstrating the implementation and use of dynamic models from an ecological modeller’s perspective who works together with field and lab ecologists. First, I will give a short overview on typical models from our own field of research, in particular aquatic ecosystems, lakes and rivers. Then, I will focus more on fundamental building blocks and modelling techniques that may be useful for other biological disciplines as well, for example: • Differential equations, e.g. Lotka-Volterra or SIR-models, • Models for laboratory systems (batch, continuous and semi-continuous cultures), • Random walk and grid-based approaches for agent-based models. It is discussed which tools are available and how dynamic models can help to improve qualitative and quantitative systems understanding. References: Soetaert, K. Petzoldt, T. & Setzer, R. W. (2010): Solving differential equations in R: package deSolve. Journal of Statistical Software 33(9), 1-25. More information can be found a the deSolve – project page: http://desolve.r-forge.r-project.org

Last modified: Dec 22, 2011, 6:30:57 PM