Loading...

Development of Optimization-Based Frameworks for the Analysis of Microbe–Microbe Interactions and Growth in Microbial Communities

Mirzaei, Soraya | 2025

0 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 58514 (02)
  4. University: Sharif University of Technology
  5. Department: Mathematical Sciences
  6. Advisor(s): Fotouhi, Morteza; Tefagh, Mojtaba
  7. Abstract:
  8. Microbial communities play key roles in maintaining human health and sustaining complex ecosystems. To gain a deeper understanding of their structure and function, it is essential to investigate the interactions and connections among microorganisms. In this study, we developed a computational approach capable of predicting metabolic interactions between microbial species pairs. Within this framework, we examine whether two species can establish competitive, parasitic, or commensal relationships over a shared extracellular metabolite. The results obtained from the proposed algorithm can be applied to identify key metabolites influencing microbial interaction patterns in natural communities, as well as in the design of targeted culture media. To evaluate the performance of this method, we applied it to well-characterized microbial communities such as the honeybee gut microbiota and leaf surface microbiota, and compared the predictions with experimental data from the phyllosphere bacterial community. Furthermore, to model microbial community growth, we introduce the MOFA algorithm, a bilevel optimization framework that simultaneously considers growth criteria at both species and community levels. By imposing constraints on species growth rates in the outer optimization, this method prevents the occurrence of forced altruism observed in previous algorithms. We evaluated the performance of MOFA both on a simplified model and on communities consisting of Desulfovibrio vulgaris and Methanococcus maripaludis, two species with a mutualistic metabolic relationship. The results demonstrated that MOFA provides more accurate predictions that align better with experimental data compared to NECom, OptCom, and Joint-FBA. In addition, for comparison with MICOM and the Microbiome Modeling Toolbox, we used datasets related to gut microbiota including co-culture experimental data, where our method consistently outperformed the available tools in most cases
  9. Keywords:
  10. Genome Scale Metabolic Model ; Bilevel Optimization Problems ; Constraint-Based Modeling ; Microbial Community Growth Rate ; Microbial Community ; Microbe–Microbe Interaction

 Digital Object List

 Bookmark

...see more