Integrating Microbiome and Clinical Data for Disease Risk Prediction in Cystic Fibrosis

This study developed a rule-based disease risk prediction model for cystic fibrosis patients that integrates microbiome composition, CFTR genetic variants, and clinical metadata using 16S rRNA sequencing data from public repositories. The model uses a baseline risk score of 50% with weighted adjustments for factors including CFTR mutations, clinical status, medications, age, and bacterial abundance to provide personalized risk assessments. Results demonstrated that therapeutic beta-lactam antibiotics significantly reduced microbiome diversity compared to subtherapeutic doses, correlating with increased exacerbation risk and highlighting the importance of preserving microbial diversity. This transparent, customizable framework advances precision medicine in cystic fibrosis management by providing clinicians with an evidence-based tool for individualized patient care.