Streptomyces species produce a vast diversity of secondary metabolites of clinical and biotechnological importance, including antibacterial that are increasingly sought after to tackle the spread of resistance. Recent developments in metabolic engineering, synthetic and systems biology have opened new opportunities to exploit Streptomyces secondary metabolism, but achieving industrial levels of production without time-consuming optimisation has remained challenging. In this thesis, we present the reconstruction and analysis of constraint-based genome-scale metabolic models to study and engineer primary and secondary metabolism of Streptomyces species. The aim of this work is to better understand so that it can aid in the increase of antibiotic production in Streptomyces spp. This would ultimately help in the discovery and production of new antibiotics to face the rise of antimicrobial resistance. This thesis starts with an introduction on Streptomyces (and other Actinobacteria) primary and secondary metabolism, and on synthetic and systems biology principles and methods relevant to Streptomyces strains engineering (Chapter I). The work presented here involved the update and validation of a high-quality genome-scale metabolic model of Streptomyces coelicolor to study its primary and secondary metabolism (Chapter II). A metabolic model of Streptomyces lividans was reconstructed and compared to S. coelicolor model to identify the metabolic differences potentially responsible for the differences in antibiotics production between and the two closely related strains (Chapter III). The comparative reconstruction and comparison method used for S. lividans was automated to compare about 50 different Actinobacteria strains and study the metabolic differences between these organisms (Chapter IV). We automated the essential process used to constrain metabolic exchanges in the metabolic model that allows condition-specific predictions, by building an R tool (Chapter V). The models and tools developed and validated in this thesis were used to help design better S. coelicolor antibiotic producing strains, by applying it to the production of actinorhodin and the heterologous production of the GE2270A antibiotic (Chapter VI). Finally, the thesis concludes with a discussion on the future applications of the models and tools developed here, as well as future developments and applications of metabolic modelling for synthetic and systems biology of antibiotic production in Streptomyces species.