Integrating Transcriptomic and Codon Usage Data for Better Therapeutic Target Characterisation of Cl

Clostridium difficile is a leading healthcare-associated illness. To find an effective treatment against this bacterium, we need a better understanding of this organism and its genotype-phenotype relationship. Genome-scale metabolic models contain all known biochemical reactions of a microorganism and can be used to study this relationship. Metabolic models are built on metabolic networks, which are composed of reactions, genes, and metabolites that have been documented for the bacterium. The standard metabolic network cannot account for the changes in this bacterium's metabolism in different environmental conditions. To address this limitation, we integrate gene expression data. To bridge the gap between gene expression data, and protein abundance, we account for the synonymous codon usage of the bacterium. To validate our metabolic model, we can compare model predictions to existing literature on C. difficile. We find that the model predicts interesting facets of the bacterium's metabolism, such as the changes in the growth of the bacterium in response to different environmental conditions. To find potential therapeutic targets using this model, we use gene essentiality and metabolic pathway sensitivity analyses.