During this seminar I will describe my research program aiming at a better understanding of the physics of supernovae and of the origin of nuclei by increasing the quality and predictive power of numerical models as well as nucleosynthesis calculations. Supernovae play essential roles in the frameworks of many branches of astrophysics: star formation, galaxy dynamics, high-energy astrophysics, galactic chemical evolution, and cosmology. In spite of their ubiquitous presence in astrophysics, there are many uncertainties related to progenitor systems, treatment of the explosions, cross section determinations at such high temperatures, and comparisons with spectra. Most popular results in the field of nucleosynthesis during explosions are still mostly based on one-spatial dimension calculations. The pioneering and very innovative aspect today is the possibility of coupling nucleosynthesis to multidimensional simulations of different type of supernovae. I will show recent results and future perspectives in multi-dimensional calculations of thermonuclear as well as core-collapse supernovae, using tracer particle method for nucleosynthesis. I will illustrate detailed comparison of 1D and 3D supernova models with nucleosynthesis calculations and discussing the needs of multi-D (and where it is needed). Despite the huge investments in nuclear physics experiments, theoretical studies establishing priority lists of reactions to be measured and precision required for astrophysics are currently very limited. During this seminar I will also discuss a priority list for future experiments and improvements in predictions of key nuclear reactions for explosive nucleosynthesis. My expertise in Galactic Chemical Evolution modelling lead to the possibility to study a dependence of the SNe yields on metallicity and their contribution over the galactic age up to reproducing the Solar System composition. During my talk I will refer different times to result of chemical evolution studies with the need of a more clear understanding of the impact of supernovae at the earliest stages of the evolution of galaxies, and their contribution to the Solar System composition. The wealth of information from galactic surveys makes this the ideal time for a theorist to understand the formation and evolution of galaxies with a new generation of chemical evolution models. To this goal, at the end of my talk, I will describe a novel project to model chemo-dynamical evolution of the cosmos, based on a N-body SPH RAMSES code making use of the framework on a moving mesh, adjusting automatically spatial resolution but using a large number of isotopes.