Different molecular representations exist to provide visual and mathematical means to depict complex structures. However, the detailing required for algorithmic applications remains a challenge. This study presents a methodology that uses two matrices to obtain a detailed molecular description. The proposed matrices are Molecule as a Graph (MAG), which describes a molecular graph providing detailed information about the bonds between atoms in a molecule, and an extended version of Structure-Oriented-Lumping (SOLex), representing each core in a molecule with 44 increments organized into vectors, capable of describing multicore structures. The combined use of these representations integrates the agility of SOLex with the structural detailing of MAG, enabling a rapid acquisition of detailed information about molecules. This can be applied in the calculation of thermodynamic properties using group contribution methods, as well as in the depiction of chemical reactions at molecular level, enabling precise identification of groups and reactive sites in molecules efficiently, even in complex algorithms. The methodology was designed for hydrocarbons from petroleum and biomass-derived sources, but can also be applied to different types of molecules, providing a versatile and efficient means of handling molecular information.