Comparative analysis of Escherichia coli genome biodiversity in adult patient or septic shock across geographical regions using WGS data
2025 (Engelska)Självständigt arbete på avancerad nivå (masterexamen), 20 poäng / 30 hp
Studentuppsats (Examensarbete)
Abstract [en]
Escherichia coli is a leading pathogen in bloodstream infections and sepsis, yet the geographic variation of sequence types, antimicrobial resistance genes and virulence factors remains unevenly described. This study aimed to address this critical knowledge gap by performing a comparative analysis of the genome biodiversity of 65 Escherichia coli isolates from adult sepsis patients across four distinct geographical regions: the United Kingdom, Spain, Australia, and Wales. Whole-genome sequencing data from 65 isolates of Escherichia coli were processed using a standardized bioinformatics pipeline enabling quality assessment, species confirmation, genome assembly, and evaluation. In-silico identification of multilocus sequence types, antimicrobial resistance genes and virulence factors was conducted. Previously published Swedish cohort results were included as a benchmark to provide a broader context for the comparative analysis. The analysis revealed significant geographical heterogeneity in the genomic profiles and resistance patterns of the isolates. A high burden of antimicrobial resistance was observed in the isolates from Wales, where the prevalence of multi-drug resistance was 75%, followed by Spain with a prevalence of 65%. In contrast, the Swedish cohort showed a substantially lower multi-drug resistance prevalence of only 6.3%. ESBL prevalence was 20% in UK, 45% in Spain, 40% in Australia, 65% in Wales, and 4% in the benchmark results from Sweden. The most frequently detected resistance genes were those conferring resistance to β-lactam, quinolone, trimethoprim, and sulphonamide antibiotics, particularly within the Welsh and Spanish datasets. Core virulence genes such as fimH, chuA, and irp2 were conserved across all isolates, though regional differences were noted in other virulence factors. Sequence typing analysis further revealed that the globally dominant high-risk clone, ST131, was highly prevalent in the Welsh dataset but was less common in other regional regions. In terms of virulence, a core set of virulence genes including fimH, chuA and irp2 was conserved across all isolates, though regional differences were noted in other virulence factors. The findings from this study underscore the dynamic and region-specific nature of Escherichia coli populations associated with sepsis. The pronounced dominance of the multi-drug resistant associated ST131 clone in Wales is a finding of significant clinical and public health concern. The results highlight that effective antibiotic stewardship and infection control strategies must be tailored to specific geographical contexts and support the urgent need for a more comprehensive, international framework for genomic surveillance of Escherichia coli and other key pathogens.
Ort, förlag, år, upplaga, sidor
2025. , s. 47
Nyckelord [en]
Federated Learning, LLaMA-3, Medical Chatbots, Model Quantization, Retrieval-Augmented Generation, AI in Healthcare
Nationell ämneskategori
Medicinsk bioinformatik och systembiologi
Identifikatorer
URN: urn:nbn:se:his:diva-25887OAI: oai:DiVA.org:his-25887DiVA, id: diva2:2002979
Ämne / kurs
Systembiologi
Utbildningsprogram
Systembiologi med inriktning mot bioinformatik - masterprogram, 120 hp
Handledare
Examinatorer
2025-10-022025-10-022025-10-02Bibliografiskt granskad