Author: Renee Tung

The power spectrum of proteins at high frequencies is remarkably well described by the flat Wilson statistics. Wilson statistics therefore plays a significant role in X-ray crystallography and more recently in cryo-EM. Specifically, modern computational methods for three-dimensional map sharpening and atomic modelling of macromolecules by single particle cryo-EM are based on Wilson statistics. In this talk we provide the first rigorous mathematical derivation of Wilson statistics. The derivation pinpoints the regime of validity of Wilson statistics in terms of the size of the macromolecule. Moreover, the analysis naturally leads to generalizations of the statistics to covariance and higher order spectra. These in turn provide theoretical foundation for assumptions underlying the widespread Bayesian inference framework for three-dimensional refinement and for explaining the limitations of autocorrelation based methods in cryo-EM.

Correlated cryo-light and cryo-electron microscopy (cryo-CLEM) has become an increasingly popular method for combining the resolving power of cryo-EM with the specificity of fluorescence. Although cryo-fluorescence microscopy suffers from optical limitations, it is a powerful way to target the resolving power of cryo-EM toward proteins of interest in heterogeneous cellular environments. Two large areas of methods developments are ongoing in both hardware and software: improving the correlation precision and accuracy to facilitate single-protein targeting, and expanding the cryo-fluorescence toolkit to include room-temperature approaches capable of targeting functions or specific protein conformations. This webinar will begin with a theoretical overview of current obstacles and considerations for software developers seeking to improve correlation precision and accuracy, and conclude with our recent efforts to improve cryo-fluorescence data quality, including characterizing the effect of cryo temperatures on fluorophores and establishing function-based localization in cryo-CLEM.

Although cryo-EM has been drastically improved to generate high-resolution three-dimensional (3D) maps that contain detailed structural information about macromolecules, the computational methods for using the data to automatically build structure models are lagging far behind. The traditional cryo-EM model building approach is template-based homology modeling. Manual de novo modeling is very time-consuming when no template model is found in the database. In recent years, de novo cryo-EM modeling using machine learning (ML) and deep learning (DL) has ranked among the top-performing methods in macromolecular structure modeling. Deep-learning-based de novo cryo-EM modeling is an important application of artificial intelligence, with impressive results and great potential for the next generation of molecular biomedicine. Accordingly, I will talk about the representative ML/DL-based de novo cryo-EM modeling methods that we developed.

Cryo electron tomograms (cryo-ET) are nothing short of amazing. There is so much to see, but wouldn’t it be nice if we could identify all those structural features in terms of their molecular identity? The combination of cryo-single molecule localization microscopy (cryo-SMLM) and cryo-ET should provide a pathway to making such identification. What are the challenges to realizing the combination’s full potential? Why the fish tank?

HEMNMA is a method to analyze continuous conformational variability of purified macromolecular complexes (in vitro), introduced in 2014 (Jin et al., Structure 22:496-506, 2014). Its software with user friendly graphical interface is part of the open-source ContinuousFlex plugin of Scipion 2 and Scipion 3 (Harastani et al., Protein Science 29:223–236, 2020). HEMNMA combines single particle image analysis, normal mode analysis, and dimension reduction techniques to visualize the full distribution of states (conformational landscape) in a low-dimensional space (usually 2D or 3D space), from which one can obtain 3D reconstructions and movies of molecular motions along desired directions. After a brief reminder on HEMNMA, I will present current developments, including combining HEMNMA with deep learning to accelerate the conformational landscape determination and an extension of HEMNMA to analyze continuous conformational variability of macromolecular complexes in cells from in situ cryo electron tomography data.     

Three-dimensional reconstruction of the electron-scattering potential of biological macromolecules from electron cryo-microscopy (cryo-EM) projection images is an ill-posed problem. The most popular cryo-EM software solutions to date rely on a regularization approach that is based on the prior assumption that the scattering potential varies smoothly over three-dimensional space. Although this approach has been hugely successful in recent years, the amount of prior knowledge that it exploits compares unfavorably with the knowledge about biological structures that has been accumulated over decades of research in structural biology. Here, a regularization framework for cryo-EM structure determination is presented that exploits prior knowledge about biological structures through a convolutional neural network that is trained on known macromolecular structures. This neural network is inserted into the iterative cryo-EM structure-determination process through an approach that is inspired by regularization by denoising. It is shown that the new regularization approach yields better reconstructions than the current state of the art for simulated data, and options to extend this work for application to experimental cryo-EM data are discussed.