Conformational dynamics is important for protein function. Biological systems can indeed adopt several conformations but it is rare that all of these conformations are known. However, using appropriate computational methods one can predict those conformations using low-resolution experimental data and a high-resolution structure. Normal mode flexible fitting (NMFF) was developed for that purpose. In NMFF a known initial high-resolution structure is deformed to fit into low-resolution data obtained from Cryo-Electron Microscopy experiments. It uses a linear combination of low-frequency normal modes in an iterative manner to deform the structure and it utilizes gradient following techniques (steepest descent method) to locally optimize the correlation coefficient between computed and measured electron density. In this work we present an extension of NMFF to experimental data derived from Small Angle Scattering (SAXS) data. Since the (two-dimensional) information content embedded in SAXS data is smaller in comparison to the (three-dimensional) cryo-EM data, the gradient procedure alone turned out to be insufficient for the optimization. Rather we implemented a trusted region method for which both gradient and curvature are used. We have also been working on developing appropriate scoring functions. Results obtained for four different proteins, Adenylate Kinase, LAO binding protein, Maltodextrin binding protein and Elongation Factor 2 will be shown.