Tesi etd-07072025-145824 |
Link copiato negli appunti
Tipo di tesi
Tesi di specializzazione (3 anni)
Autore
LOMBARDO, NADIA
URN
etd-07072025-145824
Titolo
Tecnologie di sequenziamento di nuova generazione (NGS) a supporto del controllo sugli alimenti: confronto di due piattaforme NGS nell'autenticazione di prodotti etnici venduti in Italia.
Dipartimento
SCIENZE VETERINARIE
Corso di studi
ISPEZIONE DEGLI ALIMENTI DI ORIGINE ANIMALE
Relatori
relatore Dott.ssa Giusti, Alice
correlatore Dott.ssa Gasperetti, Laura
correlatore Dott.ssa Gasperetti, Laura
Parole chiave
- metagenomica
- Next Generation Sequencing (NGS)
- sicurezza alimentare
Data inizio appello
30/07/2025
Consultabilità
Non consultabile
Data di rilascio
30/07/2028
Riassunto
Nel contesto delle attività di controllo ufficiale finalizzate all’autenticazione degli alimenti in Italia, i campioni prelevati a livello nazionale dall’Autorità Competente (AC) vengono inviati all’Istituto Zooprofilattico Sperimentale (IZS) territorialmente competente. L’analisi ufficiale per l’identificazione delle specie si basa attualmente su una tecnica validata di DNA microarray a bassa densità (MBD), efficace per un numero limitato di specie target, ma con limitazioni nell’analisi di matrici complesse contenenti specie non convenzionali o inattese, come spesso accade nei prodotti etnici. Per superare questi limiti, le tecnologie di Next Generation Sequencing (NGS) stanno emergendo come strumenti promettenti per la rilevazione di frodi in alimenti multi-specie. In questo studio, due piattaforme NGS – Illumina (ILM) e Ion Torrent (ION) – sono state impiegate, mediante approccio metabarcoding, per analizzare 56 campioni di alimenti etnici, precedentemente autenticati con MBD presso l’IZS del Lazio e della Toscana. I risultati sono stati confrontati con quelli del metodo di riferimento al fine di valutarne l’applicabilità come strumento di screening per il controllo ufficiale. Per entrambe le piattaforme è stata condotta un’analisi comparativa con MBD, valutando: 1) la concordanza qualitativa nella rilevazione delle specie (completa, parziale, assente), e 2) le metriche prestazionali relative: sensibilità relativa (rs) e tasso di overcalling (or). ILM ha mostrato una concordanza completa in 28,5% dei campioni, parziale in 55,4% e assente in 16,1%; ION ha mostrato rispettivamente 51,8%, 26,8% e 21,4%. La rs è risultata pari a 68,2% per ILM e 66,4% per ION, mentre l’or è stato del 27,5% per ILM e del 10,8% per ION. In conclusione, ILM ha evidenziato una maggiore capacità di rilevamento ma anche un rischio più elevato di sovrastime. ION, invece, ha mostrato una maggiore concordanza globale con MBD e un tasso di overcalling inferiore, a fronte di una sensibilità leggermente ridotta. Questi risultati sottolineano l’importanza di considerare congiuntamente sensibilità e specificità nell’impiego del metabarcoding per scopi regolatori. Poiché tutti i metodi analizzati forniscono dati qualitativi, l’integrazione di approcci quantitativi potrebbe rappresentare un valido supporto nel contesto del controllo ufficiale degli alimenti.
In Italy, as part of official control activities for food authentication, samples collected nationwide by the Competent Authority (CA) are sent to the territorially competent Istituto Zooprofilattico Sperimentale (IZS). The current official method for species identification is based on a validated low-density DNA microarray (LDM) technique, which is effective for a limited number of target species but shows limitations when analyzing complex matrices containing unconventional or unexpected species, as often occurs in ethnic food products. To overcome these limitations, Next Generation Sequencing (NGS) technologies are increasingly being explored as promising tools for detecting fraud in multi-species food products. In this study, two NGS platforms – Illumina (ILM) and Ion Torrent (ION) – were applied using a metabarcoding approach to analyze 56 ethnic food samples previously authenticated by LDM at the IZS of Lazio and Tuscany. The results were compared with those of the reference method to evaluate the applicability of NGS as a screening tool for official food controls. For both platforms, a comparative analysis with LDM was conducted, assessing: (1) qualitative concordance in species detection (full, partial, or none), and (2) relative performance metrics, including relative sensitivity (rs) and overcalling rate (or). ILM showed full concordance in 28.5% of samples, partial concordance in 55.4%, and no concordance in 16.1%. ION showed 51.8% full concordance, 26.8% partial, and 21.4% with no concordance. The rs was 68.2% for ILM and 66.4% for ION, while the or was 27.5% for ILM and 10.8% for ION. In conclusion, ILM demonstrated a higher species detection capacity but a greater tendency toward overestimation. Conversely, ION showed better overall agreement with LDM and a lower overcalling rate, with slightly reduced sensitivity. These results highlight the importance of jointly considering sensitivity and specificity when applying metabarcoding for regulatory purposes. Since all methods tested provide qualitative results, integrating quantitative approaches could offer valuable support within the framework of official food control.
In Italy, as part of official control activities for food authentication, samples collected nationwide by the Competent Authority (CA) are sent to the territorially competent Istituto Zooprofilattico Sperimentale (IZS). The current official method for species identification is based on a validated low-density DNA microarray (LDM) technique, which is effective for a limited number of target species but shows limitations when analyzing complex matrices containing unconventional or unexpected species, as often occurs in ethnic food products. To overcome these limitations, Next Generation Sequencing (NGS) technologies are increasingly being explored as promising tools for detecting fraud in multi-species food products. In this study, two NGS platforms – Illumina (ILM) and Ion Torrent (ION) – were applied using a metabarcoding approach to analyze 56 ethnic food samples previously authenticated by LDM at the IZS of Lazio and Tuscany. The results were compared with those of the reference method to evaluate the applicability of NGS as a screening tool for official food controls. For both platforms, a comparative analysis with LDM was conducted, assessing: (1) qualitative concordance in species detection (full, partial, or none), and (2) relative performance metrics, including relative sensitivity (rs) and overcalling rate (or). ILM showed full concordance in 28.5% of samples, partial concordance in 55.4%, and no concordance in 16.1%. ION showed 51.8% full concordance, 26.8% partial, and 21.4% with no concordance. The rs was 68.2% for ILM and 66.4% for ION, while the or was 27.5% for ILM and 10.8% for ION. In conclusion, ILM demonstrated a higher species detection capacity but a greater tendency toward overestimation. Conversely, ION showed better overall agreement with LDM and a lower overcalling rate, with slightly reduced sensitivity. These results highlight the importance of jointly considering sensitivity and specificity when applying metabarcoding for regulatory purposes. Since all methods tested provide qualitative results, integrating quantitative approaches could offer valuable support within the framework of official food control.
File
Nome file | Dimensione |
---|---|
La tesi non è consultabile. |