Publicação
Comparison of quantitative microbial risk assessment approaches using Listeria monocytogenes in Serra da Estrela cheese as a case study
| Resumo: | ABSTRACT - Whole Genome Sequencing (WGS) data has been growing and its recent application in Quantitative Microbial Risk Assessment (QMRA) has been discussed. Taking this growth into consideration, this study compared classic QMRA and WGS QMRA in terms of risk assessment questions they can answer and the way they can support decision making by risk managers. For this purpose, Listeria monocytogenes in Serra da Estrela cheese was used as a case study. Initially, a review of existing literature related to classic QMRAs and the application of WGS in QMRA was performed. This review revealed that WGS has shown advantages when integrated in the classic QMRA by allowing to fine tune each step of the risk assessment. After literature review, a classic QMRA was performed which predicted a total of 16 listeriosis cases in Portugal in one year due to the consumption of Serra da Estrela cheese. Multiple scenarios were tested, and results underline the importance of the cheese being stored at refrigeration temperatures. The WGS QMRA based on available WGS data of L. monocytogenes isolated from cheeses, using a machine learning model trained with French L. monocytogenes WGS data with known clinical frequency, predicted a clinical frequency of 37 to 54% due to Serra da Estrela cheese consumption and identified the genes and cheeses that are associated the most with clinical cases. This study concluded that both the assessed QMRA approaches are good in answering different questions and may support different types of control measures. Classic QMRA is good in giving the necessary scientific information for risk managers to decide on mitigation strategies whereas WGS QMRA allows for an early detection of outbreaks and more informed decision on product withdrawal. Therefore, this study suggests that having models to predict the clinical frequency based on WGS can be useful for risk managers as WGS data can, not only be integrated in the classic QMRA to obtain more precise results, but also be used independently as a first approach tool to promptly detect outbreaks and decide if immediate measures are required. However, further studies on the use of WGS for decision making in the risk management phase are needed for a correct use of the information. |
|---|---|
| Autores principais: | Costa, Raquel de Lobo e Oliveira |
| Assunto: | Risk assessment Whole genome sequencing Machine learning Listeria monocytogenes Cheese Avaliação do risco Whole genome sequencing Machine learning Listeria monocytogenes Queijo |
| Ano: | 2021 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório da Universidade de Lisboa |
| Resumo: | ABSTRACT - Whole Genome Sequencing (WGS) data has been growing and its recent application in Quantitative Microbial Risk Assessment (QMRA) has been discussed. Taking this growth into consideration, this study compared classic QMRA and WGS QMRA in terms of risk assessment questions they can answer and the way they can support decision making by risk managers. For this purpose, Listeria monocytogenes in Serra da Estrela cheese was used as a case study. Initially, a review of existing literature related to classic QMRAs and the application of WGS in QMRA was performed. This review revealed that WGS has shown advantages when integrated in the classic QMRA by allowing to fine tune each step of the risk assessment. After literature review, a classic QMRA was performed which predicted a total of 16 listeriosis cases in Portugal in one year due to the consumption of Serra da Estrela cheese. Multiple scenarios were tested, and results underline the importance of the cheese being stored at refrigeration temperatures. The WGS QMRA based on available WGS data of L. monocytogenes isolated from cheeses, using a machine learning model trained with French L. monocytogenes WGS data with known clinical frequency, predicted a clinical frequency of 37 to 54% due to Serra da Estrela cheese consumption and identified the genes and cheeses that are associated the most with clinical cases. This study concluded that both the assessed QMRA approaches are good in answering different questions and may support different types of control measures. Classic QMRA is good in giving the necessary scientific information for risk managers to decide on mitigation strategies whereas WGS QMRA allows for an early detection of outbreaks and more informed decision on product withdrawal. Therefore, this study suggests that having models to predict the clinical frequency based on WGS can be useful for risk managers as WGS data can, not only be integrated in the classic QMRA to obtain more precise results, but also be used independently as a first approach tool to promptly detect outbreaks and decide if immediate measures are required. However, further studies on the use of WGS for decision making in the risk management phase are needed for a correct use of the information. |
|---|