Computational models and algorithms to solve large-scale problems in Network Biology

The acquisition and measurement of biological data have dramatically changed in the recent years. In fact, High-Throughtput Sequencing (HTS) technologies are now able to capture almost all the "blueprints" that encode for life within a organism’s cell, in a parallel and rapid manner. This newly and cheaper data enabled a whole novel ensemble of approaches to unravel the mechanisms that regulate the processes in living cells. Unfortunately, these different sources of information can easily reach Terabytes of data for one single study. Consequently, in order to integrate and analyze systemically all the different data sources, novel methodologies, algorithms and software are needed to uncover the real benefits of this biological Big Data. Therefore, in my Ph.D studies I investigated those mathematical and Information Theory models that are well suited for the representation of biological phenomena using network concepts applied to large datasets. Moreover, the main idea that has driven my studies has been about making models understandable to elucidate the mechanisms of cell regulation, rather than using the most recent and very powerful deep learning approaches that may solve the problem but give you back a black box hard to dissect - i.e. making better inferences rather than robust predictions. Thus, whenever it was possible I chose simple models over complex ones. Ultimately, new models and software tools have been the results of these studies, and either research advances about new conceptual frameworks or their implementation have been published on several international conferences and journals. A full publication list is in the conclusions.