Home » training » PhD Module » Modelling Data
How can food microbiologists model their data?
Overview
The workshop is a continuation of the first workshop held during the 2nd General Meeting of BIOTRACER in Dublin. In that workshop, tutorial lectures aimed to render partners familiar with basic principles of predictive modelling. Effort was given to explain the difference between tracking and tracing models. In the current workshop, an introduction to the basic principles of predictive modelling will also be provided, whereas at a next stage, more advanced lectures will be given.
The primary objective of this workshop is to enable and/or enhance communication between data generators and data modellers. Additional objective include the training of data generators on experimental design.
More specifically, the present workshop aims to provide a better understanding of predictive modelling and train participants to model their own data. This will be achieved by combining theoretical and practical exercises in a way that participants may be self-evaluated. The latter will take place by providing raw data, which will be processed by primary and secondary models by the partners, while the outcome of modelling, e.g., final parameter estimates, will be revealed at the end of the workshop.
Outline
Dr. Panos Skandamis will cover the theoretical session of the workshop, whereas he and Dr. Antonia Gounadaki will contribute to the practical session, including practical exercises, interaction with participants and demonstrations of novel Predictive Modelling tools.
Demonstration with methodology of development and validation of predictive models will be performed, including experimental design for collection of data for modelling microbial inactivation, probability of growth and growth, classification of models (primary, secondary, tertiary), fitting primary and secondary models with linear and nonlinear regression, and validation of models by using of performance statistics and visible comparisons of predicted with observed values. The partners will receive training on the definition of dependent and independent variables as well as on the necessary data transformation in order to be further used for modelling. Demonstration on recording binary or ordinal probability of growth data will also take place, whereas there will be a special session for demonstration of Monte Carlo simulation and how it may be implemented in exposure assessment.
Proposed agenda
13.30-13.40 Classification of predictive models (Panos).
13.40-14.20 Methodology for development and validation of predictive models (Panos).
14.20-15.00 Combination of kinetic and probabilistic models (Panos)
15.00-15.15 Coffee break-Demonstration of predictive modelling software (Panos-Antonia)
15.15-15.45 Demos on the fitting of primary and secondary models to the data produced in BIOTRACER (Panos-Antonia).
15.45-16.30 Practical exercises with prototype data sets provided by the hosts of the workshop (Panos-Antonia).
Relevant literature will be provided to the attendees in advance, whereas further literature and XL files including fitted data-sets with primary and secondary models will be available after the workshop.
Pre-workshop Required Reading*
Modeling microbial responses in foods 2004 (Chapters 2 and 3), Edited by Robin C. McKellar, Xuewen Lu. Boca Raton, Fla. : CRC Press, c2004.
*Required for PhD students.
Post-workshop Reading Material
Rosso, L.; Lobry, J. R.; Bajard, S.; Flandrois, J. P., Convenient Model To Describe the Combined Effects of Temperature and pH on Microbial Growth, Applied and Environmental Microbiology, Vol. 61 Issue 2, 610-616.
Baranyi, J.; Robinson, T.P.; Kaloti, A.; Mackey, B.M., Predicting growth of Brochothrix thermosphacta at changing temperature, International Journal of Food Microbiology, Vol.27 Issue 1, 61-75.
Baranyi, József; Roberts, Terry A., A dynamic approach to predicting bacterial growth in food, International Journal of Food Microbiology, Vol.23 Issue.3-4 Part.Special Issue Predictive Modelling, 277-294.