To reach a deeper understanding of the origin of elements in the periodic table, we construct Galactic chemical evolution (GCE) models for all stable elements from C (A = 12) to U (A = 238) from first principles, i.e., using theoretical nucleosynthesis yields and event rates of all chemical enrichment sources. This enables us to predict the origin of elements as a function of time and environment. In the solar neighborhood, we find that stars with initial masses of M > 30M o˙ can become failed supernovae if there is a significant contribution from hypernovae (HNe) at M ∼ 20-50M o˙. The contribution to GCE from super-asymptotic giant branch (AGB) stars (with M ∼ 8-10M o˙ at solar metallicity) is negligible, unless hybrid white dwarfs from low-mass super-AGB stars explode as so-called Type Iax supernovae, or high-mass super-AGB stars explode as electron-capture supernovae (ECSNe). Among neutron-capture elements, the observed abundances of the second (Ba) and third (Pb) peak elements are well reproduced with our updated yields of the slow neutron-capture process (s-process) from AGB stars. The first peak elements (Sr, Y, Zr) are sufficiently produced by ECSNe together with AGB stars. Neutron star mergers can produce rapid neutron-capture process (r-process) elements up to Th and U, but the timescales are too long to explain observations at low metallicities. The observed evolutionary trends, such as for Eu, can well be explained if ∼3% of 25-50M o˙ HNe are magneto-rotational supernovae producing r-process elements. Along with the solar neighborhood, we also predict the evolutionary trends in the halo, bulge, and thick disk for future comparison with Galactic archeology surveys.