Connectome

from Wikipedia, the free encyclopedia
The white matter in a human brain, visualized by MRI .

As connectome ( English connectome ) is defined as the totality of the compounds in the nervous system of a living being . His studies on different scales dedicated to Konnektomik ( English connectomics ), a branch of neuroscience .

Since the connections of a nerve cell play a central role in determining its function, their investigation is a traditional subject of biological research. However, the pair of terms “connectome” and “connectomics” has only existed since 2005. The Human Connectome Project , approved in September 2010, contributed to its establishment, within the framework of which the National Institutes of Health researched the human connectome with a total of almost 40 million US dollars. Promote dollars.

Coining of the term

The term “connectome” goes back to the neuroscientists Olaf Sporns ( Indiana University ) and Patric Hagmann ( École polytechnique fédérale de Lausanne ), who proposed it independently in 2005.

By consciously referring to the existing terms genome and proteome , which denote the entirety of the genetic information or proteins in a living being, the name "connectome" is intended to express that the individual connections can only be understood in their mutual relationship to one another, similar to individual genes or proteins interact with one another in the organism. Similar to the proteome, the connectome is not static, but is subject to constant changes due to its neural plasticity .

The term wiring , borrowed from electrical engineering, is used for the synaptic connections between the neurons .

The connectome on different scales

Connectome research focuses on the network properties of the nervous system. The network to be examined is usually represented by a graph in the sense of graph theory , that is, as an abstract set of so-called nodes that are connected by edges. The specific meaning of the nodes and edges depends on the scale on which the nervous system is viewed. The possible orders of magnitude are roughly divided into microscale, mesoscale and macroscale.

Microscale: The connectome as a network of nerve cells

At the finest level, the microscale, a nervous system consists of neurons that are connected to one another by synapses . An examination of the nerve tissue on this scale requires its representation with a resolution in the micrometer range. In this case, the nodes in the network graph correspond to the individual neurons.

A complete reconstruction of the connectome at the micro level was first achieved in 1986 in the hermaphrodites of the nematode Caenorhabditis elegans . Your nervous system consists of 302 neurons that form around 5000 synapses, 2000 neuromuscular endplates and 600 gap junctions . It was described in 1986 on the basis of an electron microscopic examination of serial sections by a team led by biologist John White .

A similarly comprehensive investigation of more complex organisms is not possible with the current state of the art, since the number of their neurons typically runs into the billions. Based on projections, it can be assumed that the human brain, for example, contains around 10 billion neurons, which form around 100 billion synapses; In comparison, the human genome is little more than 3 billion base pairs. On this scale, it is no longer possible to manually trace all axons in successive electromicroscopic slice images. Therefore, specialized methods of machine vision are being developed for this task , but their quality has not yet quite reached that of manual segmentation . In addition, the storage and processing of the resulting amounts of data already represents a technical challenge.

Mesoscale: Cortical pillars and layers

The cerebral cortex is organized in cortical columns, groups of a few hundred or thousand neurons, with a total diameter of about 80 micrometers. They are characterized by common afferent nerve connections, are particularly strongly connected to one another and are particularly pronounced in the primary sensory areas. They are a basic processing unit of the cortex.

In addition to this vertical organization, horizontal layers can also be histologically distinguished; Due to their number, the cerebral cortex is divided into the isocortex (six layers) and the allocortex (three to five layers). While brain research has made significant advances in understanding on the micro- and macro-scale in the course of its recent history , there are still few approaches to investigate how associations of neurons work at the intermediate level.

Macro scale: connections between areas of the brain

On the basis of their anatomical properties or their function, different brain areas can be distinguished, which form the nodes of the network graph when the connectome is viewed at the macro level. These centers are connected by longer nerve fibers that make up the white matter . The general course of many large nerve bundles on the macro level is known on the basis of knowledge gained by means of special preparation and staining processes . The aim of current research (as of 2012) is to use imaging methods to examine the individual connectome of living test persons and to explain the influences of genetic factors, normal aging processes, as well as learning processes and diseases. In addition to important contributions from numerous smaller research projects, the Human Connectome Project represents the largest coordinated research program in this area between 2010 and 2015. The following three methods are currently available for research:

The diffusion imaging allows direct examination of the white matter. It measures the Brownian molecular movement of water molecules. Since this is limited by the microstructure of the nerve fibers, its preferred direction enables an estimate of the local alignment of the nerve fibers. Tractography processes use such data to algorithmically reconstruct the course of the major nerve tracts. Different methods have been proposed to derive a network graph from the diffusion data: One possibility is to take the brain areas (i.e. the nodes of the graph) from a cortical map; Gong u. a. define 78 regions in this way. A finer subdivision into 500 to 4000 nodes can be achieved by randomly grouping 8 to 64 connected voxels each, which in this case have no anatomical significance. Two nodes are connected by an edge when the tractography has reconstructed a connection between the corresponding areas; Edge weights can be derived from the number of reconstructed curves. If the common connectome is to be displayed for a whole group of test persons, an edge is only set if the connection has been reliably reproduced within the group.

In addition to diffusion imaging, whose network graphs are viewed in the context of connectomics as an expression of “structural” connectivity, functional magnetic resonance imaging (fMRI) is also used. In particular, correlations in the BOLD signal , which is an indirect expression of brain activity, are assessed as an indication of a connection between the corresponding brain regions. Measurements in the resting state are particularly common ; the networks observed are therefore referred to as resting state networks . In the context of connectomics, it is hoped that functional imaging will in particular display polysynaptic connections that contain , for example, a switch in the thalamus and are therefore not reliably recognized with diffusion imaging. In contrast to tractography methods, fMRI shows the connected brain regions, but not the course of the nerve connection itself; In return, it suffers less from crossing or fanning fiber webs, which limit the accuracy and reliability of tractography methods.

A third approach leads systematic correlations in the thickness of different cortical regions back to a connection between the correlated areas. This thickness can be estimated from conventional T1-weighted magnetic resonance tomography images . Since these correlations can only be calculated by comparing subjects, in this case there is no individual connectome, but a common graph for all subjects in the group examined.

The connectome in the context of functional specialization and integration

According to the theory of functional specialization and integration, specific functions of the brain require the integration of specialized brain areas. An essential motivation for researching the connectome is the assumption that the connections between the nerve cells contribute significantly to both aspects. For example, the function of the visual cortex is primarily determined by its afferent connection to the sensory cells in the eye. On the other hand, it is known from experiments with split-brain patients that the fiber tracts in the corpus callosum are essential for integrating visual impressions of the left visual field with the language centers .

While functional specialization through methods of electrophysiology and functional imaging has meanwhile been researched and documented relatively thoroughly, research into the connectome is expected to provide further insights into the mechanisms of functional integration, which have so far been less well understood.

literature

  • Sebastian Seung: The Connectome - Does the Circuit Diagram of the Brain Explain Our Self? Springer Spectrum, 2013, ISBN 978-3-642-34294-3 .

Web links

Commons : Connectomics  - collection of images, videos and audio files

Individual evidence

  1. ^ O. Sporns, G. Tononi, R. Kötter: The Human Connectome: A Structural Description of the Human Brain. In: PLoS Comput Biol. 1 (4), 2005, p. E42.
  2. ^ A b P. Hagmann: From Diffusion MRI to Brain Connectomics. Dissertation . École polytechnique fédérale de Lausanne, 2005.
  3. ^ JG White, E. Southgate, JN Thomson, S. Brenner: The structure of the nervous system of the nematode Caenorhabditis elegans. In: Phil. Trans. Royal Soc. London. B 314, 1986, pp. 1-340.
  4. ^ O. Sporns: Connectome. In: Scholarpedia. 5 (2), 2009, p. 5584. Revision 91162.
  5. ↑ SC Turaga, JF Murray, V. Jain, F. Roth, M. Helmstaedter, K. Briggman, W. Denk, HS Seung: Convolutional networks can learn to generate affinity graphs for image segmentation . In: Neural Computation . tape 22 , no. 2 , 2010, p. 511-538 , doi : 10.1162 / neco.2009.10-08-881 , PMID 19922289 .
  6. ↑ The Manifesto. What do brain researchers know and what can they do today? In: Brain & Mind . June 2004.
  7. ^ E. Ludwig, L. Klingler: Atlas cerebri humani. S. Karger AG, Basel 1956.
  8. ^ A b G. Gong, Y. He, L. Concha, C. Lebel, DW Gross, AC Evans, C. Beaulieu: Mapping Anatomical Connectivity Patterns of Human Cerebral Cortex Using In Vivo Diffusion Tensor Imaging Tractography . In: Cerebral Cortex . tape 19 , 2009, p. 524-536 , doi : 10.1093 / cercor / bhn102 , PMC 2722790 (free full text).
  9. ↑ a b P. Hagmann, M. Kurant, X. Gigandet, P. Thiran, VJ Wedeen, R. Meuli, J.-P. Thiran: Mapping Human Whole-Brain Structural Networks with Diffusion MRI. In: PLoS ONE . 2 (7), 2007, p. E597.
  10. ↑ KRA Van Dijk, T. Hedden, A. Venkataraman, KC Evans, SW Lazar, RL Buckner: Intrinsic Functional Connectivity As a Tool For Human Connectomics: Theory, Properties, and Optimization. In: Journal of Neurophysiology . 103, 2010, pp. 297-321.
  11. ^ Y. He, ZJ Chen, AC Evans: Small-World Anatomical Networks in the Human Brain Revealed by Cortical Thickness from MRI. In: Cerebral Cortex. 17, 2007, pp. 2407-2419.
  12. ↑ K. Friston: Functional Integration. In: K. Friston et al. (Ed.): Statistical Parametric Mapping. The Analysis of Functional Brain Images. Academic Press, 2011.
  13. ^ O. Sporns: Discovering the Human Connectome. MIT Press, 2012, p. 20.