Sustainable bio-production of fuels and chemicals is becoming cardinal in our everyday lives as resource overexploitation and pollution are increasing problems. However, bioprocesses still present numerous challenges compared to fossil alternatives. Furthermore, it is paramount to find and develop robust microorganisms that can function despite disturbances while maintaining high yields and productivity. To discover microorganisms with robust traits and to predict their industrial behaviour, we developed a mathematical method to quantify microbial robustness. The method was later used in a high-throughput setup to quantify the robustness of multiple yeast phenotypes for bioethanol production from lignocellulosic substrates. The method allowed us to discover performance and robustness trade-offs in yeast and to determine the most robust and best-performing strain for that industrial purpose. However, the physiological mechanisms and the molecular traits behind robustness have not yet been discovered. With this missing step, we collaborated with Professor Michael Desai at Harvard University (Cambridge, US). Microbial robustness is defined as the ability of a microorganism to maintain stable performance in changing environments; therefore, we hypothesized that evolving strains in constantly fluctuating environments might stimulate robust traits and consequently fixate mutations associated with robustness. The purpose of the collaboration is to identify the genetic variants of the stains that evolved in fluctuating environments and reverse-engineer those mutations to increase the robustness of strains of interest. Three Saccharomyces cerevisiae strains were evolved in three different setups. Each setup corresponded to an adaptive laboratory evolution experiment performed by sequential serial passages in 96-well plates. In two of the setups, the culture environment changed every 8 generations in one case and every 80 generations in the second case, in a total of 15 different environments. The third setup consisted of serial transfers in constant environments. Our mathematical method will quantify the robustness and fitness of the 400th-generation samples. A total of 1152 DNA samples were submitted for whole genome sequencing. We aim to link together fitness, robustness, and sequencing data to draw a map of the metabolic processes and the evolutionary mechanisms involved in microbial robustness. Ultimately, we aim to: 1. Set up a pipeline to analyse whole genome-seq data in terms of read quality control, alignment, and variant calling 2. Use the output of point 1 to compare genomes and variants of different strains, replicates, and evolution experiments 3. Link fitness and robustness data to points 1 and 2 to investigate whether there is a correlation between the fixed mutations and the manifested phenotypes Overall, this study will not only pave the way to a new methodology to increase the robustness of a given strain by adaptive laboratory evolution for industrial purposes but mainly give insight into the genetic mechanisms of microbial robustness. This discovery will also open the door to further investigation of robustness in different organisms, for example in cancer cells as they exhibit a high level of robustness against a range of therapeutic treatments.