1H4F Safer Food with Big Data
With big data analysis and machine learning algorithms we aim to identify microbiome-derived biomarkers/signatures for the early detection of potential product contamination in slaughterhouses. Through this early warning system, early interventions can take place to avoid recalls and further contamination of meat products with food pathogens. This monitoring will lead to improved food safety for consumers and reduced contamination related economical losses.
To achieve this we aim to identify such microbiome-related biomarkers within a model of pathogenic bacteria (Salmonella) that can potentially be found on the carcasses in the pig slaughter line. The expertise’s of the various partners have been used to set-up and perform several experiments to assess microbiome measurement feasibility on carcasses in the pig slaughter line and whether sample pooling strategies can be used to reduce analysis costs without losing analytical power (#1), whether there is an early warning signal in microbiome community structure associated with presence of pathogenic Salmonella (#2), whether isolated Salmonella’s harbor biofilm embracing genetic and biochemical signatures that could be exploited (#3), and whether non-microbiome collected BIG data could be used (#4).
It will provide a blue-print how to identify early-warning for potential food product contamination at the slaughterline thereby improving food safety.
This project aims to further reduce the small percentage of potential contamination using early detection signals (biomarkers) developed with innovative methods. These biomarkers derived from the existing microbiome in the production chain will act as early predictors when contamination occurs with food pathogens.
1) assess microbiome measurements feasibility on the pig slaughter line and whether sample pooling strategies can be used to reduce analysis costs without losing analytical power.
2) whether there is an early warning signal in microbiome community structure associated with the food pathogen Salmonella.
3) whether isolated Salmonella’s harbor biofilm embracing genetic / phenotypic signatures.
4) Whether other sources of non-microbiome BIG data can contribute to early warning.
- Objective 1 has been presented as preliminary data at the physical meeting in April 2019 in Utrecht, which was hosted by Vion and was finalized before the final sampling rounds in 2019.
- Objective 2 has been extended with the collection and analysis of the last half year of microbiome data completing the full year monitoring into 2020 and 2021.
- Objective 3 started after summer 2019 to be completed mostly before summer 2020. Most of the biophysical properties and genome correlation assessments are completed. Data is aggregated for publication.
- Objective 4 started early 2020 and is still ongoing trying to determine what the found associations actually mean. Will be finalized in the updated work plan 2021.
Het e-mailadres wordt niet gepubliceerd. Vereiste velden zijn gemarkeerd met *