This project was conducted during my PhD thesis: a joint collaboration between the University of Fribourg (Ecology and Informatics departments) and the HEIA-FR. Using a real dataset of niche-overlap graphs (graphs that depict the competition between species in the nature), we firstly developed a novel algorithm to detect efficiently chordless cycles and showed that these cycles are numerous in these graphs. Secondly, we investigated the structure of these networks using the theory of graphs. Finally, we simulated the abundance of species over time using systems of Lotka-Volterra differential equations and investigated the persistence of these species inside niche-overlap graphs networks using a generalized linear model.