Necrotising Soft Tissue Infections (NSTI) are a group of bacterial infections characterized by widespread tissue destruction in any layer of the human soft-tissue. These infections often result in amputation, mortality, septic shock, and a significant decline in the patient’s quality of life. Despite this, early diagnosis has proven to be challenging and approximately 50% of the patients are initially misdiagnosed owing to vague symptoms, heterogeneous patient groups, lack of an international consensus and specific diagnostic tools.
This thesis constructs a structured scientific enquiry using computational approaches to understand and explore the underlying biological mechanisms of NSTI. Data collected from variables measured in the clinic and ICU, gene expressions from human and bacterial genes, and protein and metabolite concentrations are systematically studied using modelling, data science and machine learning approaches.
This thesis reveals the regulation of bacterial responses to environmental cues, immune evasion strategies, aetiology-dependent host responses and discriminatory plasma biomarkers with potential value for diagnostic, prognostic and therapeutic approaches. Finally, this thesis expands on the intricacies in the use of network-based methods in the realm of systems medicine.