There are three major problems in structure determination that can be circumvented using the proposed method. First, currently available methods such as nuclear magnetic resonance, X-ray crystallography and cryo electron microscopy require averaging over thousands of identical molecules at least. Second, X-ray crystallography and cryo electron microscopy lead to radiation damage of the sample. Third, NMR and X-ray crystallography are limited to a certain molecular size or shape. Magnetic Resonance Imaging (MRI) is an attractive alternative since it is non destructive, has elemental contrast and is possible for single particles. However, conventional MRI has only a few microns resolution. Magnetic resonance force microscopy, the method this proposal is based on, combines the principle of MRI with scanning force microscopy and thus pushes the limits of resolution down to the nanometer range. I will add domain contrast and push the limits of resolution further by introducing partial labeling. I will test and apply this entirely new concept to tobacco mosaic viruses. That will lead to a 3D protein structure with domain contrast with nanometer resolution from a single virus particle. This method is especially promising for large, rare or heterogeneous proteins that cannot be analyzed with state of the art methods.