From organelles to cell assemblies: Creating realistic 3D models
It is well established that the shape of nerve cells, such as neurons and glial cells, can have a big impact on their functioning. Accurate structural models can help to shed more light on the relation between structure and function; more realistic structural models allow researchers to further zoom in on this question. In a new publication in Briefings in Bioinformatics, scientists from the EPFL Blue Brain Project working with an international team of scientists share a new neuroscience-dedicated framework, Ultraliser, capable of building these structural models with realistic and detailed cellular geometries that can be used in in silico neuroscience research. The software is available as open source.
While advances in imaging technologies are supplying more, and more precise, data about neurons and glial cells, creating realistic three-dimensional structural models from these large and detailed amounts of data is challenging particularly when they are needed for precise subcellular simulations. Ultraliser is designed to accomplish this objective by allowing to create accurate and biologically realistic 3D models of complex neuroscientific structures at intracellular (e.g. mitochondria and endoplasmic reticula), cellular (e.g. neurons and glia) and even multicellular scales of resolution (e.g. cerebral vasculature and minicolumns). The framework can essentially build optimized watertight mesh models and annotated volumetric models from morphological skeletons, non-watertight triangle-soups or volumetric masks of cellular, subcellular and vascular structures.
The framework represents a major leap forward in simulation-based neuroscience, making it possible to employ high-resolution 3D structural models for quantification of surface areas and volumes. “The framework has a modular and extensible architecture”, says author Marwan Abdellah, Scientific Visualization Expert at the BBP, “this makes it possible to integrate further relevant neuroscientific applications. It uses multiple data representations, for example point-and-diameter representations, low-quality three-dimensional meshes and binary volumes segmented from imaging stacks to create high fidelity structural models that can be directly plugged in detailed simulation applications.”
The power of Ultraliser is demonstrated with several use cases in which hundreds of structural models are created for potential application in diverse types of simulations. In line with Blue Brain’s continued commitment to open science, Ultraliser is publicly released. Prof. Felix Schürmann, Blue Brain Computing Director adds “The framework is unrivaled both in ease-of-use and in the accuracy of resulting geometry.”
Authors and Affiliations
Blue Brain Project, École Polytechnique Fédérale de Lausanne (EPFL), Campus Biotech, 1202 Geneva, Switzerland: Marwan Abdellah, Juan José García Cantero, Nadir Román Guerrero, Alessandro Foni, Jay S. Coggan, Eleftherios Zisis, Daniel Keller, Henry Markram & Felix Schürmann
Biological and Environmental Sciences and Engineering Division, King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia: Corrado Calì & Pierre J. Magistretti
Visual Computing Center, King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia: Marco Agus & Markus Hadwiger
Neuroscience Institute Cavalieri Ottolenghi (NICO), Orbassano, Italy: Corrado Calì
Department of Neuroscience, University of Torino, Italy: Corrado Calì
College of Science and Engineering, Hamad Bin Khalifa University, Doha, Qatar: Marco Agus
This work was supported by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research (OSR) under Award No. OSR-2017-CRG6-3438 and by funding to the Blue Brain Project, a research center of the École Polytechnique Fédérale de Lausanne (EPFL), from the Swiss government’s ETH Board of the Swiss Federal Institutes of Technology.
Abdellah, M., Cantero, J. J. G., Guerrero, N. R., Foni, A., Coggan, J. S., Calì, C., Agus, M., Zisis, E., Keller, D., Hadwiger, M., Magistretti, P. J., Markram, H., & Schürmann, F. (2023). Ultraliser: a framework for creating multiscale, high-fidelity and geometrically realistic 3D models for in silico neuroscience. Briefings in Bioinformatics 24.1 (2023). https://doi.org/10.1093/bib/bbac491