Biology subjects

Yushkevich, A.

Publications and source records attributed to Yushkevich, A..

3 recordsLinked to original sources

bioRxiv2024

Denoising, Deblurring, and optical Deconvolution for cryo-ET and light microscopy with a physics-informed deep neural network DeBCR

Computational image-quality enhancement for microscopy (deblurring, denoising, and optical deconvolution) provides researchers with detailed information on samples. Recent general-purpose deep learning solutions advanced in this task. Yet, without consideration of the underlying physics, they may yield unrealistic and non-existent details and distortions during image restoration, requiring domain expertise to discern true features from artifacts. Furthermore, the large expressive capacity of general-purpose deep learning models requires more resources to train and use in applications. We introduce DeBCR, a physics-informed deep learning model based on wavelet theory to enhance microscopy images. DeBCR is a light model with a fast runtime and without hallucinations. We evaluated the image restoration performance of DeBCR and 12 current state-of-the-art models over 6 datasets spanning crucial modalities in advanced light microscopy and cryo-electron tomography. Leveraging optic models, DeBCR demonstrates superior performance in denoising, optical deconvolution, and deblurring tasks across both LM and cryo-ET modalities.

bioinformatics↗

bioRxiv2024

Molecular architecture of synaptic vesicles.

Synaptic vesicles (SVs) store and transport neurotransmitters to the presynaptic active zone for release by exocytosis. After release, SV proteins and excess membrane are recycled via endocytosis, and new SVs are formed in a clathrin-dependent manner. This process maintains the morphology and complex molecular composition of SVs through multiple recycling rounds. Previous studies explored the molecular composition of SVs through proteomic analysis and fluorescent microscopy, proposing a model for an average SV1,2. However, the structural heterogeneity and molecular architecture of individual SVs are not well described. Here we used cryo-electron tomography to visualize morphological and molecular details of SVs isolated from mouse brains and inside cultured neurons. We describe several classes of small proteins on the SV surface and long proteinaceous densities inside SVs. We identified V-ATPases, determined a structure using subtomogram average, and showed them forming a complex with the membrane-embedded protein synaptophysin. Our bioluminescence assay revealed pairwise interactions between VAMP2 and synaptophysin and V-ATPase Voe1 domains. Interestingly, V-ATPases were randomly distributed on the surface of SVs irrespective of vesicle sizes. A subpopulation of isolated vesicles and vesicles inside neurons contained a partially assembled clathrin coat with a soccer-ball symmetry. We observed a V-ATPase under clathrin cage in several isolated clathrin-coated vesicles. Additionally, from isolated SV preparations and within hippocampal neurons we identified clathrin baskets without vesicles. We determined their preferential location in proximity to the cell membrane. Our analysis advances the understanding of individual SVs diversity and their molecular architecture.

biophysics↗

bioRxiv2023

Streamlined Structure Determination by Cryo-Electron Tomography and Subtomogram Averaging using TomoBEAR

Structures of macromolecules in their native state provide unique unambiguous insights into their functions. Cryo-electron tomography combined with subtomogram averaging demonstrated the power to solve such structures in situ at resolutions in the range of 3 Angstrom for some macromolecules. In order to be applicable to structural determination of the majority of macromolecules observable in cells in limited amounts, processing of tomographic data has to be performed in a high-throughput manner. Here we present TomoBEAR - a modular configurable workflow engine for streamlined processing of cryo-electron tomographic data for subtomogram averaging. TomoBEAR combines commonly used cryo-EM packages and reasonable presets to provide a transparent "white box" for data management and processing. We demonstrate applications of TomoBEAR to two datasets of purified proteins and to a membrane protein RyR1 in a membrane and demonstrate the ability to produce high resolution with minimal human intervention. TomoBEAR is an open-source and extendable package, it will accelerate the adoption of in situ structural biology by cryo-ET.

biophysics↗