'Changing Connectomes' by Marcus Kaiser, A book review by Brenda Walker

15th Dec 2024

BNA Associate member, Brenda Walker, shares her review of the book 'Changing Connectomes - Evolution, Development, and Dynamics in Network Neuroscience', Marcus Kaiser, MIT Press, 2020

  This hard back volume in three parts is a distant dream fulfilled for Marcus Kaiser. In 2005, when he was studying with Professor Claus C. Hilgetag, the only connectivity information to hand was from a few animals or roundworms (C.elegans), but by the time of writing, Kaiser had large data sets of ongoing human brain networks available to study from Europe and the United States, as well as the UK Biobank’s imaging studies whose twenty-year achievement was celebrated in 2022. This meant he was now able to access not only information about health and disease but also longitudinal studies of brain development during the life span, including antepartum and infancy.   

     The ‘connectome’ is defined as ‘the set of connections between neurons and regions’.  Changing Connectomes aims to ‘provide an overview of network features of the brain, how these features emerge during evolution and brain development, and how alterations ranging from normal aging to developmental and neurodegenerative disorders affect brain networks’.   Suggested readership, mentioned in the Preface, includes biomedical researchers who might be interested in neuroimaging research in the connectome development of species other than humans, those new to this area of study and computational neuroscience researchers seeking to extend their knowledge of recent developments in the field.    

    Following the detailed introduction which informs the reader about the contents of each of the fifteen chapters, the three parts are listed as covering Connectome Structure, Connectome Maturation and Connectome Changes. The volume includes an excellent glossary, thirty-nine pages of references and an index. Structure includes the features of such complex networks, their place in the evolution of neural systems together with their organisation. Maturation looks at brain development, layer formation, axonal growth, the formation of hubs, module formation and cortical folding, whereas Changes discusses development and aging, neurodevelopmental and neurodegenerative disorders together with recovery from injury and brain stimulation effects.   Information and formulae detailing results of the network studies discussed are portrayed in carefully labelled diagrams and drawings.  

    At the end of the introduction, the author addresses questions: ‘How can we determine brain connectivity?’ and ‘How can we analyse brain networks?’ To answer these, the first chapter opens with information regarding the connectome architecture, the reader being reminded that the scope, resolution and parcellation of the nervous system can influence the network features being observed.  He mentions that the scope itself could range widely when distinguishing differences between humans and other species, as human brain studies are ‘usually limited to the cortex and some subcortical structures’.  However, other areas in humans, such as the central nervous system and cerebellum can reveal a very different network as they have many more neurons.  The higher the resolution, the more nodes and areas are visible; for instance, Kaiser mentions that a human cortical network at the regional resolution that consists of 100 regions would at the resolution of individual neurons consist of 10 billion nodes. The edge density, that always corresponds, would then change from 10% at the regional resolution to 0.0001% at the neuronal resolution. He concludes by quoting research from 2008 -2018 that has shown parcellation schemes of the cortex varying from 68 regions to 360 with a ‘wide range of different anatomical, functional, or multimodal parcellation approaches’. 

    Features of Complex Networks is a long detailed chapter covering various techniques for analyses past and present, including reconstructing connectomes, the use of dyes to yield high resolution information on structural and functional connectivity, (invasive for human subjects, but used in post mortem studies). Also in use are: anatomical magnet resonance imaging scans, diffusion tensor imaging or diffusion spectrum imaging. However, caution is suggested as certain choices for establishing connectivity could be influenced by parcellation schemes and measures altering the local and global network features. Binary connections and weighted links are also explained along with global-scale aggregate measures with samples of detailed equations, topological properties of neural networks and regional scale groups of network neurons called clusters, modules, or communities.  The number of patterns involved together with the variety of networks: random, scale-free small-world, all differing from each other, are complicated but well explained.   Kaiser then considers white matter fibre tracts and available space in the brain, connection length distribution, and component place organisation, the role of long-distance connections, lateralization, and the use of topological features for network reconstruction.  A final summary highlights the challenging hope that the connectome could soon become a biomarker for diseases based on large-scale genome studies. He suggests that brain simulation studies may be the way forward, as such models can ‘go beyond the observation of patterns in the recordings of brain activity as simulated dynamics could include more complex model;’ so providing ‘simulated activity of individual neurons or local circuits which cannot be observed by noninvasive neuroimaging.’   

    The middle section of this volume describes connectivity in the light of the evolution of neural networks and the organisation of neural systems in other species as well as humans, together with the changes that appear with maturation. Also considered in depth are: patterning in brain regionalisation, organisation and refinement of synaptic activity, cell death, and the mechanisms involved in axonal growth and myelination. 

Chapter 8, entitled Formation of Hubs, reminds the reader that ‘Many real-world networks contain connected nodes called hubs. Such hubs play a crucial role for distributing or collecting information.’   Kaiser then explains their relevance in brain networks, their architecture, and the various types, locations and roles found not only in humans but also in the Macaque,  C. elegans and the cat.  In his summary, he adds factors such as birth time of nodes, the network’s accelerated growth and the degree of preferential attachment.  He also mentions genetic factors involving brain-wide gene expression atlases where ‘characteristic gene expression patterns are associated with network hubs and are conserved across species and scales.’  His conclusion suggests the role of future studies could be to assess ‘the role of different factors through simulations of brain network growth, and through comparisons with experimental data of network organizations at different developmental stages.’ 

     Chapter 9, Module Formation describes the connectivity of the maturing brain and how it develops in sequential stages:  first neurons, then dispersion, and finally the development of axons and dendrites; time dependant on types of neurons and cortical layers. When two groups of neurons connect, usually near each other spatially and often with the same cell genetic lineage, that is specifically referred to as a time window. Also discussed are the links between such time windows and the formation of modules as observed in the fruit fly, Drosophila.  Mechanisms for generating modules are discussed with a fascinating analysis of birth time neurons and time windows of C. elegans.  Many formulae are presented showing: the formation of long distance and imbalanced connections, model predictions versus neuronal connectivity in C. elegans; and network growth with three -time windows.  Comments on multiple time windows, reproducing connectivity across species and module refinement are reviewed in the chapter summary. 

       Cortical folding and its link to connectome development in primates and humans are analysed in Chapter 10 where it is mentioned that during observation of neurodevelopmental diseases, it was noted that changes in the cortical morphology correlated with changes in the folding.  The knowledge that formation of fiber tracts are linked to the brain’s folding patterns of the gyri, have encouraged the theory that ‘there are mechanistic relations between fibers and the formation of wrinkles'. Many possible changes are then discussed due to the following: folding, age and gender, disease, learning and nutrition. Mechanisms involved in generating such folding follow, including a variety of mechanical and physical models. In conclusion, the reader is reminded that actual physical space in the brain is a constraint on its organisation and connectome, and Kaiser stresses again that despite the fact that mathematics, physics and computer science can be utilised to  measure, predict, and stimulate the developmental changes of brain morphology, it is important to know that such mechanisms, that may work at the local level with regard to explaining local curvature and thickness within grey matter, may  not explain the global level of cortical folding. His hope for the future being that ‘As cortical folding and cortical thickness change for many brain disorders, a better understanding of the developmental factors might yield biomarkers for the early detection of abnormal development.’ 

    The third and final part of this volume discusses Connectome Changes looking at the concepts involved, examples of changes, and computational models to predict internal and external factors in the networks.  The author views the connectome of the brain as a fluid structure that changes as we age, as we learn, due to illness or injury or even to help reduce disease symptoms or compensate in some way, while it may also change in an attempt to regain healthy brain function.  Each of these areas are then given detailed attention in the chapters that follow ending with comments on many types of brain stimulation and the resulting factors, with models applied to different brain disorders. Marcus Kaiser’s concluding words are: 

‘Changing connectomes can mean observing how connectomes change but also actively altering their organisation or developmental trajectory.  Brain stimulation has the potential to fine-tune the wiring of the brain in the case of brain network disorders, but only if potential long-term effects and side effects can be assessed before performing an intervention. I hope that this book can be a starting point toward such applications.’ 

      References in this book abound as the author acknowledges other scientists’ contribution to research in the field of neural connectivity and other related fields. Kaiser is to be congratulated on accumulating this evidence in one dense volume to provide an overview, up to date at the time of publication, while also recounting the results of his own research in a prose style that invites the reader to share the vast recent science of the brain‘s connectomes. 

Brenda Walker.  December 2024