Current students

The main advantage of using graphs and networks for the representation of a biological system is that biological entities are involved in intricate interactions with each other to carry out various biological functions. Therefore, the need to investigate a system, considering it as a whole instead of based on independent individual components, emerges. Moreover, complex disease insurgence can be better understood through the study of mutated or dysregulated pathways in biological networks rather than individual genic mutations. The introduction of network analysis can then improve the identification of toxicogenomic disease-causing mechanisms, accurate diagnosis, treatment, and prevention The main complexity of representing a biological system in networks is the challenge of optimally fusing many interacting heterogeneous variables. Multi-Layer Networks (MLNs) present a promising and flexible framework for multilevel data representation and integration. MLNs connect and examine different molecular networks to understand how different components interact between them to determine the function of the biological system. The project has two primary objectives, namely biological inference and complex disease prediction. The first objective pertains to identifying the complexities and dysregulated pathways that influence a phenotype. The second objective concerns predicting complex diseases. The attainment of these goals requires starting with a computationally efficient strategy to connect multiple omics networks at the subcellular level with disease networks and environmental factors at the macroscopic level via MLNs. Once a comprehensive representation of each sample is introduced, graph theory and machine learning offer a strong theoretical framework for both biological inference and prediction tasks.