Publication: Detection and interpretation of cellular structures in tomograms: segmentation, localization and spatial pattern analysis
Loading...
Date
2012-02-24
Authors
Martínez-Sánchez, Antonio ; Lučić, Vladan
item.page.secondaryauthor
item.page.director
Publisher
Springer Nature
publication.page.editor
publication.page.department
DOI
https://doi.org/10.1007/978-3-031-51171-4_11
item.page.type
info:eu-repo/semantics/bookPart
Description
Abstract
A wealth of information is present in cellular cryo-ET images, from the shape of large three-dimensional cellular structures to structural details of molecular complexes. However, extracting this information is particularly challenging for complexes that cannot be identified visually and are embedded in cellular environments comprising multitude of macromolecular species that lack an obvious ordering pattern. To overcome these challenges, a number of diverse computational methods were developed for the localization, molecular identification and tracing of cellular structures. Here we review diverse computational methods that were developed for segmentation of large subcellular structures, template-based and template-free localization of complexes, and analysis of their spatial organization. We cover a wide array of method complexity, from computer-assisted human-based visualizations to machine learning and advanced algorithms, and discuss challenges posed by the recent developments of deep learning methods. New approaches for localization of molecular complexes and spatial analysis are expected to provide precise molecular maps of diverse cellular compartments, elucidate the organization of protein complexes within nanodomains and may point towards large non-periodic molecular assemblies. A wealth of information is present in cellular cryo-ET images, from the shape of large three-dimensional cellular structures to structural details of molecular complexes. However, extracting this information is particularly challenging for complexes that cannot be identified visually and are embedded in cellular environments comprising multitude of macromolecular species that lack an obvious ordering pattern. To overcome these challenges, a number of diverse computational methods were developed for the localization, molecular identification and tracing of cellular structures. Here we review diverse computational methods that were developed for segmentation of large subcellular structures, template-based and template-free localization of complexes, and analysis of their spatial organization. We cover a wide array of method complexity, from computer-assisted human-based visualizations to machine learning and advanced algorithms, and discuss challenges posed by the recent developments of deep learning methods. New approaches for localization of molecular complexes and spatial analysis are expected to provide precise molecular maps of diverse cellular compartments, elucidate the organization of protein complexes within nanodomains and may point towards large non-periodic molecular assemblies.
publication.page.subject
Citation
item.page.embargo
1-ene-2999
Collections
Ir a Estadísticas
Sin licencia Creative Commons.