On X-ray scattering model for single particles, Part II: Beyond protein crystallography

TL;DR

This paper critically examines the limitations of conventional X-ray scattering models in single-particle imaging beyond proteins, highlighting challenges, ambiguities, and the need for improved models to accurately retrieve volumetric information.

Contribution

It identifies key limitations of traditional scattering models, discusses ambiguities in density maps, and emphasizes the importance of electromagnetic boundary conditions for accurate 3D profile retrieval.

Findings

Conventional models have limitations for non-protein objects.

Ambiguities exist in interpreting density maps.

Electromagnetic boundary conditions are crucial for accurate scattering analysis.

Abstract

Emerging coherent X-ray scattering patterns of single-particles have shown dominant morphological signatures in agreement with predictions of the scattering model used for conventional protein crystallography. The key question is if and to what extent these scattering patterns contain volumetric information, and what model can retrieve it. This contribution is Part 2 out of two reports, in which we seek to clarify the assumptions of some different regimes and models of X-ray scattering and their implications for single particle imaging. In Part 1, basic concepts and existing scattering models along with their implications for nanocrystals, and also the misconception of using Diffraction Theory for volumetric scattering were addressed. Here in Part 2, specific challenges ahead of single particle imaging are addressed. Limitations of the conventional scattering model in the test case of a sphere, schemes for improving this common scattering model, ambiguities in meaning and properties of "density map" and non-classical effects, the crucial role of electromagnetic boundary conditions, the uniqueness of X-ray scattering inverse problems, and additional vulnerabilities of phase retrieval and its relation with "resolution" are discussed. We raise concerns about the unverified use of the common scattering model of protein crystallography for arbitrary objects, and also leaving fundamental questions such as meaning/uniqueness/properties of the sought 3D profiles to phase retrieval algorithms.