biostudies-literatureBlundell BBiotechnology and Biological Sciences Research Council740342https://www.ebi.ac.uk/biostudies/studies/S-EPMC9581024biostudies-literatureUnknown1Understanding the structure of a protein complex is crucial in determining its function. However, retrieving accurate 3D structures from microscopy images is highly challenging, particularly as many imaging modalities are two-dimensional. Recent advances in Artificial Intelligence have been applied to this problem, primarily using voxel based approaches to analyse sets of electron microscopy images. Here we present a deep learning solution for reconstructing the protein complexes from a number of 2D single molecule localization microscopy images, with the solution being completely unconstrained. Our convolutional neural network coupled with a differentiable renderer predicts pose and derives a single structure. After training, the network is discarded, with the output of this method being a structural model which fits the data-set. We demonstrate the performance of our system on two protein complexes: CEP152 (which comprises part of the proximal toroid of the centriole) and centrioles.Frontiers in bioinformatics3D Structure From 2D Microscopy Images Using Deep Learning.PMC9581024BB/S507519/1BB/M009513/1Sieben CManley SCox SBlundell BCh'ng QRosten Efalse3D Structure From 2D Microscopy Images Using Deep Learning.Understanding the structure of a protein complex is crucial in determining its function. However, retrieving accurate 3D structures from microscopy images is highly challenging, particularly as many imaging modalities are two-dimensional. Recent advances in Artificial Intelligence have been applied to this problem, primarily using voxel based approaches to analyse sets of electron microscopy images. Here we present a deep learning solution for reconstructing the protein complexes from a number of 2D single molecule localization microscopy images, with the solution being completely unconstrained. Our convolutional neural network coupled with a differentiable renderer predicts pose and derives a single structure. After training, the network is discarded, with the output of this method being a structural model which fits the data-set. We demonstrate the performance of our system on two protein complexes: CEP152 (which comprises part of the proximal toroid of the centriole) and centrioles.2021-01-01T00:00:00Z20212025-04-03T21:29:33.756Z2025-04-03T21:29:33.756ZS-EPMC95810243630374110.3389/fbinf.2021.740342