Correlation of Light- and Electron Microscopic Images with Convolutional Neural Networks
Rick Seifert, Sebastian M. Markert, Sebastian Britz, Veronika Perschin, Christoph Erbacher, Christian Stigloher, Philip Kollmannsberger
Center for Computational and Theoretical Biology, University of Würzburg, Würzburg, Germany
Abstract
In correlative light and electron microscopy (CLEM), automated correlation-based alignment is not directly possible due to the different contrast of EM and fluorescence images. Registration is thus often done by hand or semi-automatically using fiducial markers. We developed “DeepCLEM”, a fully automated CLEM registration workflow: a convolutional neural network predicts the fluorescent signal from the EM images, which is then automatically registered to the experimentally measured chromatin signal from the sample using correlation-based alignment. The complete workflow is available as a Fiji macro and, while targeted at array tomography, can be adapted for other imaging modalities as well as for 3D stacks.