@article{ToulkeridouGutierrezBaumetal.2023, author = {Toulkeridou, Evropi and Gutierrez, Carlos Enrique and Baum, Daniel and Doya, Kenji and Economo, Evan P.}, title = {Automated segmentation of insect anatomy from micro-CT images using deep learning}, volume = {3}, journal = {Natural Sciences}, number = {4}, doi = {10.1002/ntls.20230010}, year = {2023}, abstract = {Three-dimensional (3D) imaging, such as micro-computed tomography (micro-CT), is increasingly being used by organismal biologists for precise and comprehensive anatomical characterization. However, the segmentation of anatomical structures remains a bottleneck in research, often requiring tedious manual work. Here, we propose a pipeline for the fully-automated segmentation of anatomical structures in micro-CT images utilizing state-of-the-art deep learning methods, selecting the ant brain as a test case. We implemented the U-Net architecture for 2D image segmentation for our convolutional neural network (CNN), combined with pixel-island detection. For training and validation of the network, we assembled a dataset of semi-manually segmented brain images of 76 ant species. The trained network predicted the brain area in ant images fast and accurately; its performance tested on validation sets showed good agreement between the prediction and the target, scoring 80\% Intersection over Union (IoU) and 90\% Dice Coefficient (F1) accuracy. While manual segmentation usually takes many hours for each brain, the trained network takes only a few minutes. Furthermore, our network is generalizable for segmenting the whole neural system in full-body scans, and works in tests on distantly related and morphologically divergent insects (e.g., fruit flies). The latter suggests that methods like the one presented here generally apply across diverse taxa. Our method makes the construction of segmented maps and the morphological quantification of different species more efficient and scalable to large datasets, a step toward a big data approach to organismal anatomy.}, language = {en} } @inproceedings{HarthBastTroidletal.2023, author = {Harth, Philipp and Bast, Arco and Troidl, Jakob and Meulemeester, Bjorge and Pfister, Hanspeter and Beyer, Johanna and Oberlaender, Marcel and Hege, Hans-Christian and Baum, Daniel}, title = {Rapid Prototyping for Coordinated Views of Multi-scale Spatial and Abstract Data: A Grammar-based Approach}, booktitle = {Eurographics Workshop on Visual Computing for Biology and Medicine (VCBM)}, doi = {10.2312/vcbm.20231218}, year = {2023}, abstract = {Visualization grammars are gaining popularity as they allow visualization specialists and experienced users to quickly create static and interactive views. Existing grammars, however, mostly focus on abstract views, ignoring three-dimensional (3D) views, which are very important in fields such as natural sciences. We propose a generalized interaction grammar for the problem of coordinating heterogeneous view types, such as standard charts (e.g., based on Vega-Lite) and 3D anatomical views. An important aspect of our web-based framework is that user interactions with data items at various levels of detail can be systematically integrated and used to control the overall layout of the application workspace. With the help of a concise JSON-based specification of the intended workflow, we can handle complex interactive visual analysis scenarios. This enables rapid prototyping and iterative refinement of the visual analysis tool in collaboration with domain experts. We illustrate the usefulness of our framework in two real-world case studies from the field of neuroscience. Since the logic of the presented grammar-based approach for handling interactions between heterogeneous web-based views is free of any application specifics, it can also serve as a template for applications beyond biological research.}, language = {en} } @article{LongrenEigenShubitidzeetal.2023, author = {Longren, Luke L. and Eigen, Lennart and Shubitidze, Ani and Lieschnegg, Oliver and Baum, Daniel and Nyakatura, John A. and Hildebrandt, Thomas and Brecht, Michael}, title = {Dense Reconstruction of Elephant Trunk Musculature}, volume = {33}, journal = {Current Biology}, doi = {10.1016/j.cub.2023.09.007}, pages = {1 -- 8}, year = {2023}, abstract = {The elephant trunk operates as a muscular hydrostat and is actuated by the most complex musculature known in animals. Because the number of trunk muscles is unclear, we performed dense reconstructions of trunk muscle fascicles, elementary muscle units, from microCT scans of an Asian baby elephant trunk. Muscle architecture changes markedly across the trunk. Trunk tip and finger consist of about 8,000 extraordinarily filigree fascicles. The dexterous finger consists exclusively of microscopic radial fascicles pointing to a role of muscle miniaturization in elephant dexterity. Radial fascicles also predominate (at 82\% volume) the remainder of the trunk tip and we wonder if radial muscle fascicles are of particular significance for fine motor control of the dexterous trunk tip. By volume, trunk-shaft muscles comprise one-third of the numerous, small radial muscle fascicles, two-thirds of the three subtypes of large longitudinal fascicles (dorsal longitudinals, ventral outer obliques, and ventral inner obliques), and a small fraction of transversal fascicles. Shaft musculature is laterally, but not radially, symmetric. A predominance of dorsal over ventral radial muscles and of ventral over dorsal longitudinal muscles may result in a larger ability of the shaft to extend dorsally than ventrally and to bend inward rather than outward. There are around 90,000 trunk muscle fascicles. While primate hand control is based on fine control of contraction by the convergence of many motor neurons on a small set of relatively large muscles, evolution of elephant grasping has led to thousands of microscopic fascicles, which probably outnumber facial motor neurons.}, language = {en} } @article{KiewiszBaumMuellerReichertetal.2023, author = {Kiewisz, Robert and Baum, Daniel and M{\"u}ller-Reichert, Thomas and Fabig, Gunar}, title = {Serial-section electron tomography and quantitative analysis of the microtubule organization in 3D-reconstructed mitotic spindles}, volume = {13}, journal = {Bio-protocol}, number = {20}, doi = {10.21769/BioProtoc.4849}, year = {2023}, language = {en} } @article{SchmittTitschackBaum2024, author = {Schmitt, Kira and Titschack, J{\"u}rgen and Baum, Daniel}, title = {CoDA: Interactive Segmentation and Morphological Analysis of Dendroid Structures Exemplified on Stony Cold-Water Corals}, arxiv = {http://arxiv.org/abs/2406.18236}, year = {2024}, abstract = {Dendroid stony corals build highly complex colonies that develop from a single coral polyp sitting in a cup-like skeleton, called corallite, by asexual reproduction, resulting in a tree-like branching pattern of its skeleton. Despite their beauty and ecological importance as reef builders in tropical shallow-water reefs as well as in cold-water coral mounds in the deep ocean, systematic studies investigating the ontogenetic morphological development of such coral colonies are largely missing. One reason for this is the sheer number of corallites - up to several thousands in a single coral colony. Another limiting factor, especially for the analysis of dendroid cold-water corals, is the existence of many secondary joints in the ideally tree-like structure that make a reconstruction of the skeleton tree extremely tedious. Herein, we present CoDA, the Coral Dendroid structure Analyzer, a visual analytics suite that allows for the first time to investigate the ontogenetic morphological development of complex dendroid coral colonies, exemplified on three important framework-forming dendroid cold-water corals: Lophelia pertusa (Linnaeus, 1758), Madrepora oculata (Linnaeus, 1758), and Goniocorella dumosa (Alcock, 1902). Input to CoDA is an initial instance segmentation of the coral polyp cavities (calices), from which it estimates the skeleton tree of the colony and extracts classical morphological measurements and advanced shape features of the individual corallites. CoDA also works as a proofreading and error correction tool by helping to identify wrong parts in the skeleton tree and providing tools to quickly correct these errors. The final skeleton tree enables the derivation of additional information about the calices/corallite instances that otherwise could not be obtained, including their ontogenetic generation and branching patterns - the basis of a fully quantitative statistical analysis of the coral colony morphology. Part of CoDA is CoDA.Graph, a feature-rich link-and-brush user interface for visualizing the extracted features and 2D graph layouts of the skeleton tree, enabling the real-time exploration of complex coral colonies and their building blocks, the individual corallites and branches. In the future, we expect CoDA to greatly facilitate the analysis of large stony corals of different species and morphotypes, as well as other dendroid structures, enabling new insights into the influence of genetic and environmental factors on their ontogenetic morphological development.}, language = {en} } @article{VohraHerreraTavhelidseSucketal.2024, author = {Vohra, Sumit Kumar and Herrera, Kristian and Tavhelidse-Suck, Tinatini and Knoblich, Simon and Seleit, Ali and Boulanger-Weill, Jonathan and Chambule, Sydney and Aspiras, Ariel and Santoriello, Cristina and Randlett, Owen and Wittbrodt, Joachim and Aulehla, Alexander and Lichtman, Jeff W. and Fishman, Mark and Hege, Hans-Christian and Baum, Daniel and Engert, Florian and Isoe, Yasuko}, title = {Multi-species community platform for comparative neuroscience in teleost fish}, journal = {bioRxiv}, doi = {10.1101/2024.02.14.580400}, year = {2024}, abstract = {Studying neural mechanisms in complementary model organisms from different ecological niches in the same animal class can leverage the comparative brain analysis at the cellular level. To advance such a direction, we developed a unified brain atlas platform and specialized tools that allowed us to quantitatively compare neural structures in two teleost larvae, medaka (Oryzias latipes) and zebrafish (Danio rerio). Leveraging this quantitative approach we found that most brain regions are similar but some subpopulations are unique in each species. Specifically, we confirmed the existence of a clear dorsal pallial region in the telencephalon in medaka lacking in zebrafish. Further, our approach allows for extraction of differentially expressed genes in both species, and for quantitative comparison of neural activity at cellular resolution. The web-based and interactive nature of this atlas platform will facilitate the teleost community's research and its easy extensibility will encourage contributions to its continuous expansion.}, language = {en} } @misc{HajarolasvadiBaum2024, author = {Hajarolasvadi, Noushin and Baum, Daniel}, title = {Data for Training the DeepOrientation Model: Simulated cryo-ET tomogram patches}, doi = {10.12752/9686}, year = {2024}, abstract = {A major restriction to applying deep learning methods in cryo-electron tomography is the lack of annotated data. Many large learning-based models cannot be applied to these images due to the lack of adequate experimental ground truth. One appealing alternative solution to the time-consuming and expensive experimental data acquisition and annotation is the generation of simulated cryo-ET images. In this context, we exploit a public cryo-ET simulator called PolNet to generate three datasets of two macromolecular structures, namely the ribosomal complex 4v4r and Thermoplasma acidophilum 20S proteasome, 3j9i. We select these two specific particles to test whether our models work for macromolecular structures with and without rotational symmetry. The three datasets contain 50, 150, and 450 tomograms with a voxel size of 10 ̊A, respectively. Here, we publish patches of size 40 × 40 × 40 extracted from the medium-sized dataset with 26,703 samples of 4v4r and 40,671 samples of 3j9i. The original tomograms from which the samples were extracted are of size 500 × 500 × 250. Finally, it should be noted that the currently published test dataset is employed for reporting the results of our paper titled "DeepOrientation: Deep Orientation Estimation of Macromolecules in Cryo-electron tomography" paper.}, language = {en} } @article{LuetzkendorfMatkovicRachidLiuetal.2025, author = {L{\"u}tzkendorf, Janine and Matkovic-Rachid, Tanja and Liu, Sunbin and G{\"o}tz, Torsten and Gao, Lili and Turrel, Oriane and Maglione, Marta and Grieger, Melanie and Putignano, Sabrina and Ramesh, Niraja and Ghelani, Tina and Neumann, Alexander and Gimber, Niclas and Schmoranzer, Jan and Stawrakakis, Anastasia and Brence, Blaž and Baum, Daniel and Ludwig, Kai and Heine, Martin and Mielke, Thorsten and Liu, Fan and Walter, Alexander and Wahl, Markus and Sigrist, Stephan}, title = {Blobby is a synaptic active zone assembly protein required for memory in Drosophila}, volume = {16}, journal = {Nature Communications}, doi = {10.1038/s41467-024-55382-9}, year = {2025}, language = {en} } @article{YangKnoetelCiecierskaHolmesetal.2024, author = {Yang, Binru and Kn{\"o}tel, David and Ciecierska-Holmes, Jana and W{\"o}lfer, Jan and Chaumel, J{\´u}lia and Zaslansky, Paul and Baum, Daniel and Fratzl, Peter and Dean, Mason N.}, title = {Growth of a tessellation: geometric rules for the development of stingray skeletal patterns}, volume = {11}, journal = {Advanced Science}, number = {48}, doi = {10.1002/advs.202407641}, year = {2024}, language = {en} } @article{MayerBaumAmbellanetal.2024, author = {Mayer, Julius and Baum, Daniel and Ambellan, Felix and von Tycowicz, Christoph and for the Alzheimer's Disease Neuroimaging Initiative,}, title = {Shape-based Disease Grading via Functional Maps and Graph Convolutional Networks with Application to Alzheimer's Disease}, volume = {24}, journal = {BMC Medical Imaging}, doi = {10.1186/s12880-024-01513-z}, year = {2024}, abstract = {Shape analysis provides methods for understanding anatomical structures extracted from medical images. However, the underlying notions of shape spaces that are frequently employed come with strict assumptions prohibiting the analysis of incomplete and/or topologically varying shapes. This work aims to alleviate these limitations by adapting the concept of functional maps. Further, we present a graph-based learning approach for morphometric classification of disease states that uses novel shape descriptors based on this concept. We demonstrate the performance of the derived classifier on the open-access ADNI database differentiating normal controls and subjects with Alzheimer's disease. Notably, the experiments show that our approach can improve over state-of-the-art from geometric deep learning.}, language = {en} }