Hatzimanikatis, VassilyMiskovic, LjubisaAndreozzi, Stefano2015-06-162015-06-16201510.5075/epfl-thesis-6558https://infoscience.epfl.ch/handle/20.500.14299/115141urn:nbn:ch:bel-epfl-thesis6558-7Metabolic Engineering is fostered by an increasing effort and progress in the collection of large and accurate 'omics' datasets. The abundance of available measurements and advances in measurement techniques only alleviate to a certain extent the uncertainty around the physiological states of an organism. In fact, due to intrinsic complexity of metabolic networks, it is not possible to determine their exact intracellular states by integrating only the currently available fluxomics, metabolomics and kinetic data into a model. For these reasons, it is desirable to have methods that further reduce the uncertainty in depicting the actual intracellular states. To tackle this problem, in this thesis we propose novel computational approaches, based on Monte Carlo sampling and machine learning classification, that reduce this uncertainty.enmetabolic engineeringuncertaintymetabolic networksMonte Carlo Samplingmachine learningstatistical classificationenzymatic reactionsmetabolomicsfluxomicssamplingthesis::doctoral thesis