Abstract
Advancements in cryo-electron tomography (cryoET) allow the structure of macromolecules to be determined in situ, which is crucial for studying membrane protein structures and their interactions in the cellular environment. However, membranes are often highly curved and have a strong contrast in cryoET tomograms, which masks the signals from membrane proteins. These factors pose difficulties in observing and revealing the structures of membrane proteins in situ. Here, we report a membrane-flattening method and the corresponding software, MPicker, designed for the visualization, localization, and orientation determination of membrane proteins in cryoET tomograms. This method improves the visualization of proteins on and around membranes by generating a flattened tomogram that eliminates membrane curvature and reduces the spatial complexity of membrane protein analysis. In MPicker, we integrated approaches for automated particle picking and coarse alignment of membrane proteins for sub-tomogram averaging. MPicker was tested on tomograms of various cells to evaluate the method for visualizing, picking, and analyzing membrane proteins.