Brain network analysis: a practical tutorial

Abstract

How are our brains wired? How are pathways between neurons organized? What patterns of connections allow us to think the way we do, or distinguish our ways of thinking from those of other animals? These and related questions are the bread and butter of an excellent new textbook. Fundamentals of Brain Network Analysis by Fornito, Zalesky and Bullmore, is a thorough and didactic presen-tation of the tools available to research scientists wishing to engage in the emerging field of network neuroscience (Bullmore and Sporns, 2009). Blending computational tools and mathematical frameworks from physics, engineer-ing, statistics, and computer science with the reams of data now being collected from diverse neural systems, network neuroscience is a truly interdisciplinary and ground-break-ing field poised to transform our understanding of the brain. Rather than focusing solely on the function of single neurons or brain regions, these efforts expand the purview of our interests to the pattern of interactions be-tween neural elements, suggesting that these patterns can offer significant insight into the workings of the mind (Fig. 1). Indeed, a network or graph representation—in which component parts are represented as network nodes, and in which the interaction between two nodes is represented by an edge—offers a mathematical construct for studying, predicting, and controlling exactly these patterns (Bollobas, 1985). Because of its transdisciplinary nature as well as its rela-tive infancy, network neuroscience has remained a challen-ging topic for professors to teach and students to learn. The difficulty largely lies in the lack of a quintessential text. While previous efforts have either focused on network fun-damentals (Newman, 2010) or on conceptual applications of these ideas to neuroimaging data (Sporns, 2010, 2012), no text has previously combined both disciplines in an ac-cessible and comprehensive 'how-to'. This beautiful tome therefore fills a much-needed gap, offering the first and only textbook-style presentation of the network-based tools that have proven useful in addressing specific questions and hypotheses in the neurosciences. The book will undoubtedly serve as a critical resource for both stu-dents and teachers alike as network neuroscience becomes a ubiquitous topic in graduate education in the neurosciences. The book is co-written by three seminal scientists in the field, who each lend their expertise and knowledge to make this piece a particularly rich read. Indeed, their collective background in psychology, psychiatry, and engineering per-fectly complements the interdisciplinary nature of the topics that they cover. While remaining true to the mathematical complexity and theoretical depth of the ideas (Estrada and Knight, 2016), the authors are careful to consistently point out the relevance of these tools for understanding cognition and behaviour (Medaglia et al., 2015), and their alteration in neurological disorders and psychiatric disease (Stam, 2014). In addition, the authors' eclectic interests expand the scope of the work to include topics from invertebrates to vertebrates (van den Heuvel et al., 2016), from molecu-lar biology to genetics, and from motor control to learning. To organize this expansive content, the book walks step-by-step through the building blocks that compose a net-work, and how these building blocks can be extracted