A proteins function is intimately related to its structure whether it is in solution or a membrane. However, determining a protein's secondary and higher order structure quickly and reproducibly can be a challenge. The environment of the protein and even its concentration, as in the case of high concentration biopharmaceutical products, make the task even more challenging due to interference in any data collected from non-protein parts of the sample. Protein spectroscopic data contain information about the protein's secondary and also tertiary structure. However, extracting 'the answer' from a spectrum is often challenging. We have addressed this issue by collecting circular dichroism, infra red absorbance, and Raman data on a wide range of proteins and developed a self-organising map approach to the data analysis of unknown proteins using references sets of data from proteins of known structure. This opens the way to simple analysis of known proteins such as biopharmaceutical products and new proteins discovered in a wide variety of research projects. The focus of the talk will be on illustrating how useful the methodology can to answer important questions for e.g. biopharmaceuticals, plant allergens, and common human proteins. To apply such methodologies to membrane proteins one had to be aware of the artefacts that occur during data collection. For membrane peptides and proteins and protein assemblies and fibres, one of the challenges with UV spectroscopies, such as electronic circular dichroism, is the light scattering that contributes artefacts to the spectrum that is measured. Before implementing any structure fitting approach the light scattering needs to be removed. This will also be addressed.