Abstract:
Infectious diseases, exceeding political and geographical boundaries, persist as significant health threats that burden populations globally. Since ancient times, they have consistently challenged the growth and welfare of all living organisms, including humans. Additionally, the emergence of antimicrobial resistance complicates disease management, diminishing the efficacy of conventional treatments and presenting notable challenges in prevention and control. Recent viral pandemics, such as COVID-19, highlight the urgent need for effective worldwide preparedness and new therapeutic strategies. In response to these challenges, computational systems biology and in silico modeling have emerged as powerful assets in combating infectious diseases. Genotype-driven prediction of cellular phenotypes is essential to understand how genetic variations influence disease outcomes, guide personalized treatments, and identify potential drugtargets. In this regard, constraint-based modeling methods have evolved to facilitate a mechanistic understanding of metabolic physiology based on imposed constraints and experimental data. This thesis uses systems biology to study the genotype–phenotype relationship of cellular metabolism. More specifically, it explores the applications of constraint-based modeling across multiple projects, aiming to address different infections and antimicrobial-resistant pathogens. The primary focus of the first project is the antiviral target prediction, leveraging systems biology and constraint-based modeling. This project involves the reconstruction and analysis of cell-specific host-virus metabolic models to detect exploitable inhibitory pathways. The inhibitory efficacy of the identified compounds was also computationally validated across multiple known variants. Nevertheless, the rise in complexity of mathematical models raises problems in quality control and reproducibility. The second project focuses on model standardization and documentation, aiming to enhance the reusability and interoperability of computational models. It presents a Python-based annotation software to automate the assignment of systems biology ontology terms by inferring expert knowledge from model structures. The third project explores the metabolic modeling of human nasal microbiota by focusing on a particular nosocomial pathogenic member named Acinetobacter baumannii. The construction of the first collection of metabolic networks for various multidrug-resistant A. baumannii strains is presented.Their analysis yields valuable insights into strain-specific metabolic capabilities and vulnerabilities, that could guide the development of novel antimicrobial strategies targeting metabolicpathways. Finally, the last project investigates the metabolic phenotypes of Rothia mucilaginosa, an opportunistic pathogen with a multifaceted role in health and disease. The integration of experimental data alongside the construction of the first manually curated metabolic network aim to unravel potential metabolic aspects that could serve as focal points for future research endeavors.The research presented in this thesis signifies meaningful strides in infectious disease intervention via computational systems biology and metabolic modeling. Through a combination of modeling, data analysis, experiments, and software development, this thesis effectively offers comprehensive insights into previously unexplored areas, thereby advancing the understanding and application of systems biology principles in various fields.
Simulation