Determining the conformation of a small molecule inside a huge molecular weight structure is crucial for the understanding of the fundamental molecular processes that drive the interactions. Many of these complexes are neither crystalline nor soluble and are thus difficult to study by classical methods such as NMR, X-ray diffraction etc. To the best of our knowledge, a limited number of atomic structures of such systems were reported to date. In our project, we propose a new strategy based on the synergetic combination of organic synthesis (chimio-, regio- and stereo-specific tritium labeling; carbon-13 and nitrogen-15 labeling), solid state NMR and molecular modeling as a novel unique tool-kit to determine the conformation of a small molecule embedded in a high molecular and non-crystalline assembly. To develop our strategy we choose a model small molecule, the Phe-Phe dipeptide that forms either crystals or self-assembled nanotubes depending on the solvent. If the crystalline atomic structure of Phe-Phe has been solved, the structure of the self-assembled nanotubes of Phe-Phe is still unknown. To solve such structure, precise intra- and intermolecular distances should be determined to get not only the conformation of the molecule but also its packing within the assembly. We will develop our strategy (chemical synthesis, solid state NMR and molecular modeling) on Phe-Phe crystal. This strategy will be then applied to determine the atomic structure of Phe-Phe nanotubes. The result of this project is expected to pave the way for numerous forthcoming applications such as pharmacology, biology (determination of a ligand structure bounded to its receptor, self-assembled molecules) and nanotechnology (determination of the conformation and the precise position of a small molecule within a supramolecular architecture).