Towards Community Driven Food Safety Model Repositories

Abstract

Transferring predictive microbial models from research into real world food manufacturing or risk assessment applications is still a challenge for members of the food safety modelling community. Such knowledge transfer could be facilitated if publicly available food safety model repositories would exist.

This research therefore aimed at identification of missing resources hampering the establishment of community driven food safety model repositories. Existing solutions in related scientific disciplines like Systems Biology and Data Mining were analyzed.

On the basis of this analysis, some factors which would promote the establishment of community driven model repositories were identified – among them: a standardized information exchange format for models and rules for model annotation. As a consequence a proposal for a Predictive Modelling in Food Markup Language (PMF-ML) together with a prototypic implementation on the basis of the Systems Biology Markup Language (SBML) has been developed. In addition the adoption of MIRIAM guidelines for model annotation is proposed. In order to demonstrate the practicability of the proposed strategy, existing predictive models previously published in the scientific literature were re-implemented using an open source software tool called PMM-Lab. The models are made publicly available in the first community Food Safety Model Repository called openFSMR (https://sites.google.com/site/openfsmr/).

This work illustrates that a standardized information exchange format for predictive microbial models can be established by adoption of resources from Systems Biology. Harmonized description and annotation of predictive models will also contribute to increased transparency and quality of food safety models.