Section: New Results

3D Convolutional Neural Networks for macromolecule localization in cryo-electron tomograms of intact cells

Participants : Emmanuel Moebel, Charles Kervrann.

Figure 9. Experimental tomogram of a Chlamydomonas Reinhardtii cell (central slice) and the 3D segmentation obtained by our convolutional neural network. The network is able to identify 3 macromolecule classes (membrane-bound ribosome, cytoplasmic ribosome and proteasome) as well as the cell membrane (tomogram by courtesy of Max-Planck Institute, Martinsried, Germany).

In this study, we focus on macromolecule localization and classification in cryo-electron tomography images. Cryo-electron tomography (cryo-ET) allows one to capture 3D images of cells in a close to native state, at sub-nanometer resolution. However, noise and artifact levels are such that heavy computational processing is needed to access the image content. We propose a deep learning framework to accurately and jointly localize multiple types and states of macromolecules in cellular cryo-electron tomograms. We compare this framework to the commonly-used template matching method on both synthetic and experimental data. On synthetic image data, we show that our framework is very fast and produces superior detection results. On experimental data, the detection results obtained by our method correspond to an overlap rate of 86% with the expert annotations, and comparable resolution is achieved when applying subtomogram averaging. In addition, we show that our method can be combined to template matching procedures to reliably increase the number of expected detections. In our experiments, this strategy was able to find additional 20.5% membrane-bound ribosomes that were missed or discarded during manual annotation.

Collaborators: Damien Larivière (Fondation Fourmentin-Guilbert),

                          Julio Ortiz, Antonio Martinez (Max-Planck Institute, Martinsried, Germany).