Today, more than 40 million people worldwide are affected by neurodegenerative disorders. Onset of these diseases is associated with insoluble fibrillar protein aggregates, termed amyloids. The molecular origin and the link between amyloid formation and disease aetiology are unclear and there are not are available therapies for these disorders. Strong evidence links propensity of proteins to misfolding and aggregation to the pathological biology implicated in the onset of these diseases. Despite its importance, unraveling amyloids properties and formation is still a formidable experimental challenge, mainly because of their nanoscale dimensions and their heterogeneous and transient nature. Therefore, the investigation of the misfolding of monomers and oligomers into fibrils and their mechanical and structural properties is central to understand their stability and toxicity in the body and to design new therapeutic strategies to the amyloid diseases problem. The main objective of this PhD thesis was the biophysical investigation of amyloids structure and formation at the single scale aggregate. This objective was pursed mainly by the use of both conventional and innovative Atomic Force Microscopy (AFM) methodologies, such as peak force quantitative nanomechanical mapping (PF-QNM) and infrared nanospectroscopy (nanoIR). These methods were assessed to resolve the complex and heterogeneous energy landscape of proteins aggregation and to provide direct information on aggregates properties. Initially, we used AFM to compare the kinetics of aggregation of wild type and mutated forms of huntingtin and a-synuclein. In the first case, we focused on the effects of N-terminal post-translation modifications in the fibrillization. We demonstrated that a phosphorylation of the protein significantly slowed down the aggregation. In the latter case, we investigated wild type and H50Q mutated form of a-synuclein. We demonstrated the strict link between the disease and the mutation, which enhanced aggregation. Successively, we studied the early stages of a-synuclein fibrillization and we showed that the amyloid assembly proceed directly through the formation of single monomeric strands, which hierarchically assembly into the mature fibrils. Moreover, we performed force spectroscopy experiments, which confirmed the non-mature structure of this species and enabled studying their force of interaction with surfaces. Successively, PF-QNM was applied to investigate the mechanical properties of single aggregates forming during fibrillization of amyloid proteins. We demonstrated that β-sheet content is a major factor determining their intrinsic stiffness. Finally, nanoIR was applied to investigate at the nanoscale the misfolding process and the structure of the species present during the aggregation. The technique proved to be ideal to characterize individually the amyloid species formed by the Josephin domain of ataxin-3. For the first time, we were able to link their nanomechanical properties and secondary structure at the nanoscale. Innovative AFM-based techniques enabled correlating morphological and ultrastructural properties of amyloids at the single aggregate scale. Thus, they represent a future fruitful avenue to unravel protein misfolding process and the mechanisms of amyloid formation. The comprehension of these fundamental processes could allow the design of pharmacological approaches to contrast the onset of amyloid diseases.