Small angle scattering (SAS) provides critical structural information about biological macromolecules and nanomaterials in solution, but extracting 3D structural information from 1D scattering data remains challenging. DENSS is an algorithm that reconstructs electron density maps directly from solution scattering data using an iterative projection approach with minimal assumptions, avoiding many limitations of conventional bead-modeling methods. DENSS can analyze virtually any particle type, including soluble proteins, membrane proteins, RNA/DNA complexes, and nanoparticles. Since its initial publication, DENSS has been substantially expanded with new capabilities that enhance both the accuracy and applicability of density reconstructions from SAS data. Recent developments include robust support for molecular symmetry (cyclic, dihedral, and icosahedral), enabling higher resolution reconstructions of symmetric assemblies. An integrated indirect Fourier transform approach provides improved data fitting with rigorous error propagation. Enhanced alignment and averaging algorithms significantly reduce computational time while improving map quality. DENSS now includes tools for calculating theoretical scattering profiles from atomic models with superior accuracy compared to existing methods, facilitating direct comparison between experimental densities and atomic structures. Ongoing developments expand DENSS capabilities to address increasingly complex structural biology problems, including methods for analyzing conformational changes, guided reconstructions incorporating partial structural information, and decomposition of heterogeneous samples. These advances enable investigations of dynamic systems, transient states, and mixture characterization that are difficult or impossible with conventional approaches. DENSS is freely available as open-source Python code (github.com/tdgrant1/denss) and through a web server (denss.ccr.buffalo.edu), making advanced SAS analysis accessible to the broader structural biology community. This presentation will demonstrate current capabilities and preview emerging features that further establish density-based analysis as a powerful complement to traditional SAS modeling approaches.
Grant et al. (Sun,) studied this question.