Keynote Presentation: SuperResNET: Learning in-cell macromolecular architecture from SMLM data w/ Live Q&A

Protein structures are now being resolved at the atomic level, but deciphering their molecular organization in the cell remains a challenge. Super-resolution microscopy enables the use of fluorescent-based molecular localization tools to study molecular structure at the nanoscale level in the intact cell, bridging the mesoscale gap to classical structural biology methodologies. SuperResNET is an integrated machine learning-based analysis software for visualizing and quantifying 3D point cloud data acquired by single molecule localization microscopy (SMLM). The computational modules of SuperResNET include correction for multiple blinking of a single fluorophore, denoising, segmentation (clustering), and feature extraction, which are then used for cluster group identification, modularity analysis, blob retrieval and visualization in 2D and 3D. More recent updates to SuperResNET allow two-channel interaction distance analysis to determine how two proteins interact within macromolecular assemblies. SuperResNET can be effectively and easily applied to any SMLM event list from which it rapidly learns macromolecular architecture in the intact cell. I will describe the ability of network graph analysis software (SuperResNET) to determine molecular structure from dSTORM and MinFlux single molecule localization microscopy. Use cases to be described include molecular analysis of the nucleopore complex, structural changes to clathrin coated pits by inhibitors of clathrin endocytosis and the identification and structural characterization of caveolae and non-caveolar caveolin-1 scaffolds.

Learning Objectives:

1. Explain how SMLM breaks the diffraction barrier

2. Compare structure determination by SuperResNET to that obtained by cryoEM

3. Describe the structure and function of Cav1 scaffolds and caveolae