TY - THES A1 - Herter, Felix T1 - Supervised Classification of Microtubule Ends: An Evaluation of Machine Learning Approaches N2 - Aim of this thesis was to evaluate the performance of three popular machine learning methods – decision trees, support vector machines, and neural networks – on a supervised image classification task from the domain of cell biology. Specifically, the task was to classify microtubule ends in electron tomography images as open or closed. Microtubules are filamentous macromolecules of the cytoskeleton. Distribution of their end types is of interest to cell biologists as it allows to analyze microtubule nucleation sites. Currently classification is done manually by domain experts, which is a difficult task due to the low signal-to-noise ratio and the abundance of microtubules in a single cell. Automating this tedious and error prone task would be beneficial to both efficiency and consistency. Images of microtubule ends were obtained from electron tomography reconstructions of mitotic spindles. As ground truth data for training and testing four independent expert classifications for the same samples from different tomograms were used. Image information around microtubule ends was extracted in various formats for further processing. For all classifiers we considered how the performance varies when different preprocessing techniques (per-feature and per-image standardization) are applied. or decision trees and support vector machines we also evaluated the effect of training on a) imbalanced versus under- and over-sampled data and b) image-based vs feature-based input for specifically designed features. The results show that for decision trees and support vector machines classification on features outperforms classification on images. Both methods give most equalized per-class accuracies when the training data was undersampled and when preprocessed with per-image standardization prior to features extraction. Neural networks gave the best results when no preprocessing was applied. The final decision tree, support vector machine, and neural network obtained accuracies on the test set for (open,closed ) samples of (62%, 72%), (66%, 70%), and (61%, 78%) respectively, when considering all samples where at least one expert assigned a label. Restricting the test set to samples with at least three agreeing expert labels raised these to (78%, 84%), (74%, 92%), and (82%, 88%). It can be observed that many samples misclassified by the algorithms were also difficult to classify for the experts. Y1 - 2018 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:0297-zib-68395 ER - TY - GEN A1 - Redemann, Stefanie A1 - Lantzsch, Ina A1 - Lindow, Norbert A1 - Prohaska, Steffen A1 - Srayko, Martin A1 - Müller-Reichert, Thomas T1 - A switch in microtubule orientation during C. elegans meiosis N2 - In oocytes of many organisms, meiotic spindles form in the absence of centrosomes [1–5]. Such female meiotic spindles have a pointed appearance in metaphase with microtubules focused at acentrosomal spindle poles. At anaphase, the microtubules of acentrosomal spindles then transition to an inter- chromosomal array, while the spindle poles disappear. This transition is currently not understood. Previous studies have focused on this inter- chromosomal microtubule array and proposed a pushing model to drive chromosome segregation [6, 7]. This model includes an end-on orientation of microtubules with chromosomes. Alternatively, chromosomes were thought to associate along bundles of microtubules [8, 9]. Starting with metaphase, this second model proposed a pure lateral chromosome-to-microtubule association up to the final meiotic stages of anaphase. Here we applied large-scale electron tomography [10] of staged C. elegans oocytes in meiosis to analyze the orientation of microtubules in respect to chromosomes. We show that microtubules at metaphase I are primarily oriented laterally to the chromosomes and that microtubules switch to an end-on orientation during progression through anaphase. We further show that this switch in microtubule orientation involves a kinesin-13 microtubule depolymerase, KLP-7, which removes laterally associated microtubules around chromosomes. From this we conclude that both lateral and end-on modes of microtubule-to-chromosome orientations are successively used in C. elegans oocytes to segregate meiotic chromosomes. T3 - ZIB-Report - 18-34 Y1 - 2018 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:0297-zib-69855 SN - 1438-0064 ER - TY - JOUR A1 - Redemann, Stefanie A1 - Lantzsch, Ina A1 - Lindow, Norbert A1 - Prohaska, Steffen A1 - Srayko, Martin A1 - Müller-Reichert, Thomas T1 - A switch in microtubule orientation during C. elegans meiosis JF - Current Biology N2 - In oocytes of many organisms, meiotic spindles form in the absence of centrosomes [1–5]. Such female meiotic spindles have a pointed appearance in metaphase with microtubules focused at acentrosomal spindle poles. At anaphase, the microtubules of acentrosomal spindles then transition to an inter- chromosomal array, while the spindle poles disappear. This transition is currently not understood. Previous studies have focused on this inter- chromosomal microtubule array and proposed a pushing model to drive chromosome segregation [6, 7]. This model includes an end-on orientation of microtubules with chromosomes. Alternatively, chromosomes were thought to associate along bundles of microtubules [8, 9]. Starting with metaphase, this second model proposed a pure lateral chromosome-to-microtubule association up to the final meiotic stages of anaphase. Here we applied large-scale electron tomography [10] of staged C. elegans oocytes in meiosis to analyze the orientation of microtubules in respect to chromosomes. We show that microtubules at metaphase I are primarily oriented laterally to the chromosomes and that microtubules switch to an end-on orientation during progression through anaphase. We further show that this switch in microtubule orientation involves a kinesin-13 microtubule depolymerase, KLP-7, which removes laterally associated microtubules around chromosomes. From this we conclude that both lateral and end-on modes of microtubule-to-chromosome orientations are successively used in C. elegans oocytes to segregate meiotic chromosomes. Y1 - 2018 U6 - https://doi.org/10.1016/j.cub.2018.07.012 SN - 0960-9822 ER - TY - CHAP A1 - Lindow, Norbert A1 - Redemann, Stefanie A1 - Brünig, Florian A1 - Fabig, Gunar A1 - Müller-Reichert, Thomas A1 - Prohaska, Steffen T1 - Quantification of three-dimensional spindle architecture T2 - Methods in Cell Biology Part B N2 - Mitotic and meiotic spindles are microtubule-based structures to faithfully segregate chromosomes. Electron tomography is currently the method of choice to analyze the three-dimensional (3D) architecture of both types of spindles. Over the years, we have developed methods and software for automatic segmentation and stitching of microtubules in serial sections for large-scale reconstructions. 3D reconstruction of microtubules, however, is only the first step toward biological insight. The second step is the analysis of the structural data to derive measurable spindle properties. Here, we present a comprehensive set of techniques to quantify spindle parameters. These techniques provide quantitative analyses of specific microtubule classes and are applicable to a variety of tomographic reconstructions of spindles from different organisms. Y1 - 2018 U6 - https://doi.org/10.1016/bs.mcb.2018.03.012 SN - 0091-679X VL - 145 SP - 45 EP - 64 PB - Academic Press ER - TY - JOUR A1 - Fabig, Gunar A1 - Kiewisz, Robert A1 - Lindow, Norbert A1 - Powers, James A. A1 - Cota, Vanessa A1 - Quintanilla, Luis J. A1 - Brugués, Jan A1 - Prohaska, Steffen A1 - Chu, Diana S. A1 - Müller-Reichert, Thomas T1 - Sperm-specific meiotic chromosome segregation in C. elegans JF - eLife Y1 - 2020 U6 - https://doi.org/10.7554/eLife.50988 VL - 9 SP - e50988 ER - TY - JOUR A1 - Lindow, Norbert A1 - Brünig, Florian A1 - Dercksen, Vincent J. A1 - Fabig, Gunar A1 - Kiewisz, Robert A1 - Redemann, Stefanie A1 - Müller-Reichert, Thomas A1 - Prohaska, Steffen A1 - Baum, Daniel T1 - Semi-automatic stitching of filamentous structures in image stacks from serial-section electron tomography JF - bioRxiv N2 - We present a software-assisted workflow for the alignment and matching of filamentous structures across a 3D stack of serial images. This is achieved by combining automatic methods, visual validation, and interactive correction. After an initial alignment, the user can continuously improve the result by interactively correcting landmarks or matches of filaments. Supported by a visual quality assessment of regions that have been already inspected, this allows a trade-off between quality and manual labor. The software tool was developed to investigate cell division by quantitative 3D analysis of microtubules (MTs) in both mitotic and meiotic spindles. For this, each spindle is cut into a series of semi-thick physical sections, of which electron tomograms are acquired. The serial tomograms are then stitched and non-rigidly aligned to allow tracing and connecting of MTs across tomogram boundaries. In practice, automatic stitching alone provides only an incomplete solution, because large physical distortions and a low signal-to-noise ratio often cause experimental difficulties. To derive 3D models of spindles despite the problems related to sample preparation and subsequent data collection, semi-automatic validation and correction is required to remove stitching mistakes. However, due to the large number of MTs in spindles (up to 30k) and their resulting dense spatial arrangement, a naive inspection of each MT is too time consuming. Furthermore, an interactive visualization of the full image stack is hampered by the size of the data (up to 100 GB). Here, we present a specialized, interactive, semi-automatic solution that considers all requirements for large-scale stitching of filamentous structures in serial-section image stacks. The key to our solution is a careful design of the visualization and interaction tools for each processing step to guarantee real-time response, and an optimized workflow that efficiently guides the user through datasets. Y1 - 2020 U6 - https://doi.org/10.1101/2020.05.28.120899 ER - TY - GEN A1 - Lindow, Norbert A1 - Redemann, Stefanie A1 - Fabig, Gunar A1 - Müller-Reichert, Thomas A1 - Prohaska, Steffen T1 - Quantification of Three-Dimensional Spindle Architecture N2 - Mitotic and meiotic spindles are microtubule-based structures to faithfully segregate chromosomes. Electron tomography is currently the method of choice to analyze the three-dimensional architecture of both types of spindles. Over the years, we have developed methods and software for automatic segmentation and stitching of microtubules in serial sections for large-scale reconstructions. Three-dimensional reconstruction of microtubules, however, is only the first step towards biological insight. The second step is the analysis of the structural data to derive measurable spindle properties. Here, we present a comprehensive set of techniques to quantify spindle parameters. These techniques provide quantitative analyses of specific microtubule classes and are applicable to a variety of tomographic reconstructions of spindles from different organisms. T3 - ZIB-Report - 18-07 Y1 - 2018 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:0297-zib-66562 SN - 1438-0064 ER - TY - GEN A1 - Lindow, Norbert A1 - Brünig, Florian A1 - Dercksen, Vincent J. A1 - Fabig, Gunar A1 - Kiewisz, Robert A1 - Redemann, Stefanie A1 - Müller-Reichert, Thomas A1 - Prohaska, Steffen T1 - Semi-automatic Stitching of Serial Section Image Stacks with Filamentous Structures N2 - In this paper, we present a software-assisted workflow for the alignment and matching of filamentous structures across a stack of 3D serial image sections. This is achieved by a combination of automatic methods, visual validation, and interactive correction. After an initial alignment, the user can continuously improve the result by interactively correcting landmarks or matches of filaments. This is supported by a quality assessment that visualizes regions that have been already inspected and, thus, allows a trade-off between quality and manual labor. The software tool was developed in collaboration with biologists who investigate microtubule-based spindles during cell division. To quantitatively understand the structural organization of such spindles, a 3D reconstruction of the numerous microtubules is essential. Each spindle is cut into a series of semi-thick physical sections, of which electron tomograms are acquired. The sections then need to be stitched, i.e. non-rigidly aligned; and the microtubules need to be traced in each section and connected across section boundaries. Experiments led to the conclusion that automatic methods for stitching alone provide only an incomplete solution to practical analysis needs. Automatic methods may fail due to large physical distortions, a low signal-to-noise ratio of the images, or other unexpected experimental difficulties. In such situations, semi-automatic validation and correction is required to rescue as much information as possible to derive biologically meaningful results despite of some errors related to data collection. Since the correct stitching is visually not obvious due to the number of microtubules (up to 30k) and their dense spatial arrangement, these are difficult tasks. Furthermore, a naive inspection of each microtubule is too time consuming. In addition, interactive visualization is hampered by the size of the image data (up to 100 GB). Based on the requirements of our collaborators, we present a practical solution for the semi-automatic stitching of serial section image stacks with filamentous structures. T3 - ZIB-Report - 19-30 Y1 - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:0297-zib-73739 SN - 1438-0064 ER - TY - JOUR A1 - Weber, Britta A1 - Greenan, Garrett A1 - Prohaska, Steffen A1 - Baum, Daniel A1 - Hege, Hans-Christian A1 - Müller-Reichert, Thomas A1 - Hyman, Anthony A1 - Verbavatz, Jean-Marc T1 - Automated tracing of microtubules in electron tomograms of plastic embedded samples of Caenorhabditis elegans embryos JF - Journal of Structural Biology Y1 - 2012 UR - http://www.sciencedirect.com/science/article/pii/S1047847711003509 U6 - https://doi.org/10.1016/j.jsb.2011.12.004 VL - 178 IS - 2 SP - 129 EP - 138 ER - TY - CHAP A1 - Weber, Britta A1 - Möller, Marit A1 - Verbavatz, Jean-Marc A1 - Baum, Daniel A1 - Hege, Hans-Christian A1 - Prohaska, Steffen T1 - Fast Tracing of Microtubule Centerlines in Electron Tomograms T2 - BioVis 2011 Abstracts, 1st IEEE Symposium on Biological Data Visualization Y1 - 2011 ER -