Speaker: Parth Singhal
Topic: cryoDRGN based heterogeneity analysis of cryo-EM datasets.
Abstract: Knowledge of molecular structures and their conformations is essential to the understanding of biological functions. In the past few years, cryo-EM single-particle imaging has been quite helpful for the reconstruction of 3D protein volumes from 2D cryo-EM images, but these 2D images contain particle snapshots belonging to different conformations and not just one protein structure. Recent developments in the scope of deep generative machine learning models have given rise to Variational Autoencoders. One of the recent software packages, cryoDRGN (Zhong et al.), uses this powerful scheme for the 3D reconstruction of protein structures and their conformations. During the presentation, I will be talking about the use of cryoDRGN to analyze heterogeneity in real and simulated cryo-EM datasets along with the use of Machine Learning methods to generate Free Energy Landscapes and their further comparative analysis.