Chronic wounds cost the NHS £8.2 billion annually, and bacterial infections precede 90% of amputations. Our group first demonstrated skin microbiome as an important healing outcome predictor in chronic wounds. However, our understanding of their mechanisms is limited due to poor translation of murine models and lack of in-vitro models fully representing human skin microbiome. We address this key knowledge gap by developing a human ex-vivo skin microbiome platform to dissect the crucial role of bacteria in wound pathology.  We have curated a biobank of over 300 Staphylococcus strains from healthy skin and wound infections. We compared biofilm forming capabilities, antibiotic resistance and haemolytic activity. Each characteristic was graded; isolates with highest and lowest grades (most to least virulent) were selected for testing strain diversity in our human ex-vivo wound model.   Our preliminary data reveals strains of same species varying distinctly in biofilm formation, antibiotic resistance, and haemolytic potential. Notably, enhanced biofilm formation was observed under low pH and high glucose conditions. Antibiotic resistance was distributed primarily across S. epidermidis and S. aureus, with strain-specific resistance patterns. Strain-level functional variations were evident in our ex-vivo wound model, with divergent effects on wound closure rates, linked to bacterial virulence and isolation site. Ongoing long-read whole genome sequencing analysis is identifying genomic variations within strains, which will link to donor metadata, phenotypic characteristics and wound healing outcomes.   These findings demonstrate our ex-vivo model as predictive for bacterial pathogenicity screening and highlight importance of strain-level analysis in understanding microbe-host interactions in wound healing.