Integration of Bioinformatics and Machine Learning to characterize Fusobacterium nucleatum’s pathogenicity

Fusobacterium nucleatum has been found to be associated with cancer lesions in both oral and colon cancers. Although important studies have dissected the clinical aspects of its remarkable pathogenicity, there is a lack of molecular studies. Our computational work based on bioinformatics and machine learning methodologies has predicted potential pathogenicity islands. The study has involved the analysis of genome-based compositional bias, promoter maps, codon adaptation index, protein structure, and characterized these genomic regions on the basis of predicting base compositional bias, promoter mapping, protein abundances, and interactions, metabolic model to characterize these regions. Although most of the currently in use pathogenicity islands finder software detect the presence of three pathogenicity islands, our analysis suggests that only one is present. Furthermore, we have investigated and discussed the metabolic advantages of pathogenicity, particularly iron ion scavenging activity. Our work has two immediate and important benefits: the improved understanding of the biological processes that shape the pathogenicity and evolution of Fusobacterium nucleatum at the molecular level and the improved ability of integrating and automating the state-of-the-art bioinformatics tools and machine learning approaches in the inference of the mechanistic interpretability of a pathogenic phenotype.