Cajal Blue Brain
Any man can be, if he wants, the sculptor of his own brain
Santiago Ramón y Cajal
The Cajal Blue Brain Project (CBBP) was approved in 2009 for a period of 10 years (until 2018). This project has made it possible to create a multidisciplinary team of more than 50 researchers (anatomists, physiologists, mathematicians and computer scientists). As a result of the CBBP, several tools and new computational methods have been developed that represent an important technological contribution to the study of the brain.
Making an Impact in the Study of the Brain
The Cajal Blue Brain project is the Spanish contribution to the Blue Brain Project, an international approach designed to create a functional brain model by means of reverse engineering of the mammalian brain using the Blue Gene supercomputer from IBM.
Motivation
One of the main goals of neuroscience is to understand the biological mechanisms responsible for human mental activity. In particular, the study of the cerebral cortex is and without any doubt will be the greatest challenge for science in the next centuries, since it represents the foundation of our humanity. In other words, the cerebral cortex is the structure whose activity is related to the capabilities that distinguish humans from other mammals. Thanks to the development and evolution of the cerebral cortex we are able to perform highly complex and specifically human tasks, such as writing a book, composing a symphony or developing technologies.
For these reasons the Blue Brain project emerged in 2005, when the L’Ecole Polytechnique Fédérale de Lausanne (Switzerland) and IBM jointly launched an ambitious project to create a functional brain model by means of reverse engineering of the mammalian brain, using the Blue Gene supercomputer from IBM. The aim was to understand the functioning and dysfunction of the brain through detailed simulations. By late 2006, the Blue Brain project had created a model of the basic functional unit of the brain, the neocortical column. However, the goals set by the project, which covered a period of 10 years, imposed its conversion into an international initiative (The Blue Brain Project, Nat Rev Neurosci. 7, 153-160, 2006). In this context, the Cajal Blue Brain project, the Spanish contribution to this international project, started in January 2009 led by the Universidad Politécnica de Madrid (UPM) and the Consejo Superior de Investigaciones Científicas (CSIC) .
Neuronal forest simulation
Art and Technical Direction: Luis Pastor, Ángel Rodríguez, Susana Mata and Sofía Bayona – Art and Technical – Production and Development: Juan Pedro Brito and Luis Miguel Serrano – Technical Advice: José Miguel Espadero – Art Advice: Eva Cortés – Scientific Advice: Javier DeFelipe y Ruth Benavides-Piccione
“The garden of neurology offers the investigator captivating spectacles and incomparable artistic emotions. In it, my aesthetic instincts were at last full satisfied. Like the entomologist hunting for brightly colored butterflies, my attention was drawn to the flower garden of the gray matter that contained cells with delicate and elegant forms, the mysterious butterflies of the soul, the beating of whose wings may some day (who knows?) clarify the secret of mental life. […] Even from the aesthetic point of view, the nervous tissue contains the most charming attractions. In our parks is there any tree more elegant and luxurious than the Purkinje cell from the cerebellum or the psychic cell, that is the famous cerebral pyramid?”
From the connectome to the synaptome
Based on the idea that most connections are established by chemical point-to-point synapses, the terms ‘connectome’ and ‘synaptome’ have been proposed to facilitate the description of the maps of connections at different levels of resolution. The term connectome can be used to refer to maps at the macroscopic and mesoscopic levels, which also allows putative synaptic contacts to be mapped, while synaptome refers to the map of true synaptic contacts at the ultrastructural level (From the connectome to the synaptome: an epic love story, Science 330:1198-1201, 2010). Electron microscopy with serial section reconstruction is the gold standard method for tracing the connections. However, obtaining long series of sections is rather time-consuming and challenging. Consequently, the reconstruction of large tissue volumes is usually impossible.
The introduction of automated or semi-automated electron microscopy techniques at the turn of the century represented a major advance in the study of the synaptome as long series of consecutive sections can now be obtained with little user intervention. As this technology becomes more popular, it will have a huge impact on the study of the ultrastructure of the brain. Despite these high hopes, the principal drawback is that complete reconstructions of whole brains are only possible in some invertebrates or for relatively simple nervous systems, whereas for small mammals like the mouse, it is impossible to fully reconstruct the brain at the ultrastructural level. This is because the magnification needed to visualize and classify the synaptic junctions (i.e., excitatory and inhibitory) and to measure their sizes and shapes accurately enough yields relatively small images (in the order of tens of μm2). As a result, it is only possible to obtain incomplete synaptomes. It seems clear that only by combining studies at the macro-, meso-, and nano-scopic levels can we fully understand the structural arrangement of the brain as a whole.
Integration of microanatomical data. Schematic representation to show how we could deal with the problem of imprecise connectomes and incomplete synaptomes focusing on the cortical columns. A. Instead of reconstructing all cellular components within the column, the principles governing the structural design of cells can be obtained by using data from a few 3D reconstructed neurons and applying mathematical tools to determine the statistical structure of the neurons to computationally synthesize model neurons. The cells can be labeled with markers that allow full visualization of their dendritic and axonal arbors and then 3D reconstructed at the light microscope level, allowing the morphometric analysis of individual cells (i.e., patterns of dendritic arbors, distribution and density of dendritic spines, etc.). These data are also critical for modeling neuronal function such as synaptic integration in dendrites and dendritic spines. For instance, based on the dendritic spine distribution and their morphology (different colors represent different sizes), it is possible to generate maps of putative synaptic currents. B. Another set of structural data comes from measurements of the grey matter thickness, the volume fraction of cortical elements (neuropil, neurons, glia and blood vessels), neuron and glia density per volume, together with the patterns of local (intralaminar, translaminar) and long-range (cortico-cortical, thalamo-cortical, cortico-thalamic, subcortical extra-thalamic) connections. To determine the synaptic contribution of pyramidal cells in a given cortical layer, it is impractical to reconstruct all of these cells at the electron microscope level. Instead, this parameter could be inferred by combining quantitative light microscopy data on the total number and microanatomical characteristics of these cells on the one hand, with the average density of axo-spinous and axo-dendritic synapses obtained by analyzing multiple samples of the 3D reconstructed neuropil, using automated electron microscopy techniques and tools for Image analysis, segmentation, and quantification of different types of synapses (green, asymmetric synapses; red, symmetric synapses).
Taken from Neuroanatomy and Global Neuroscience. Neuron. 2017 95:14-18, 2017.
Developing algorithms to reconstruct synaptic connections
It is important to emphasize that acquiring multiple samples at different scales (light and electron microscopy) allows us to obtain a dataset that can be statistically analyzed in search of general patterns of organization (Reconstruction and Simulation of Neocortical Microcircuitry. Cell 163:456-492, 2015). This multiple sampling approach assures unprecedented accuracy, since we obtain both precise quantitative data and statistical variability information. The data can be used to identify common and differing principles of organization and to develop algorithms to reconstruct synaptic connections for use in brain models (Figure Integration of microanatomical data).
Furthermore, it seems that the most appropriate approach to make neuroanatomical studies more significant is to link detailed structural data with the incomplete light and electron microscopy wiring diagrams and integrate this neuroanatomical information with genetic, molecular and physiological data. This integration would allow the generation of models that present the data in a form that can be used to reason, make predictions and suggest new hypotheses to discover new aspects of the structural and functional organization of the brain (e.g., Reconstruction and Simulation of Neocortical Microcircuitry. Cell 163:456-492, 2015).
One of the strengths of the Cajal Blue Brain project is that all the participating laboratories and research groups will be coordinated, so that all the effort will be channelled towards a specific objective, using strictly common methodological criteria. Thus, the data generated in a laboratory can be effectively used by other research groups. Definitively, the Cajal Blue Brain project is structured in such a way that it will work as a single, large multidisciplinary laboratory. In this way, the project will generate significant advances in our understanding of the structure and function of the normal brain.
Laboratory Cajal Cortical Circuits
Centro de Tecnología Biomédica, Universidad Politécnica de Madrid
Campus de Montegancedo s/n
Pozuelo de Alarcón 28223 (Madrid) Spain
Tel: +34 910679250
