Algorithms for atomic-resolution structure identification of nanocrystals

Division Of Basic Sciences - Nci

Investigators

Linked publications & trials

Abstract

Ensembles of synthesized ligand-protected nanocrystals, each with a unique atomic structure, can be imaged tumbling in solution using Graphene-Liquid Cell Electron Microscopy (GLC-EM). Projection images of differently rotated nanocrystals are acquired using a direct electron detector with high temporal ( 2.5 ms) resolution. The analysis of time-series data of nanocrystals imaged in solution to obtain 3D structural information is an emerging field of research with great potential for improving our understanding of the physiochemical properties of nanocrystals, which thermodynamically deviate from the expectations derived from the bulk material. We previously developed a "one-particle 3D reconstruction method" based on imaging of individual Pt nanocrystals rotating in solution and the use of Single-Particle Analysis (SPA) to obtain an ensemble of 3D reconstructions (Science 2015 & 2020). However, the standard SPA workflow cannot straightforwardly be applied to reconstruct atomic- resolution 3D density maps of the nanocrystals. A number of critical steps need to be addressed: (1) the individual nanocrystals need to be accurately tracked throughout the time-series and robust tracking algorithms have proven difficult to develop, (2) the strong interfering background of the GLC needs to be subtracted, (3) low-quality images that result from the nanocrystals moving vertically out of the narrow depth of focus of the aberration-corrected microscope need to be identified and rejected and (4) tailored strategies for 2D and 3D alignment and averaging that differ from those used in biological cryo-EM are needed.