Computational Systems Biology & Study of structure and function in cultured neuronal networks
General objective: This research line studies the relationship between structure, dynamics and function in neuronal networks using, as an experimental model, primary cultures of neurons from locust (Schistocerca gregaria) ganglia and it aims to understand how the neural network is capable of initiating an activity, how it responds to failures on its connections or what the relationship is between physiology and function/dysfunction in small cultured neuronal networks.
Detailed description: The study of how an assembly of isolated neurons self-organizes to form a complex neural network is a fundamental problem to be addressed. Previous studies highlighted that the organization of the neuronal network before reaching its mature state is not random, being instead characterized by a high clustering and short paths.
In vitro primary cultures of dissociated invertebrate neurons from locust ganglia are used to investigate the morphological evolution of assemblies of living neurons, as they self-organize from collections of separated cells into elaborated, modular, networks.
In particular, we are developing a complete software for the identification of neurons and neurites location, able to ultimately extract an adjacency matrix from each image of the culture. This, on its turn, will allow us to perform statistical analyses of some relevant network topological observables at different stages of the culture’s development, and to quantify the main characteristics of a generic assembly of isolated neurons when it self-organizes to form a complex neural network.
We observe that in-vitro neural networks life cycle involves three different processes interplaying at the same time:
1) From the initial culture until 5-6 Days In-Vitro (DIV), the main process is the growth process, characterized by growth of dendrites and axons, and fattening of the soma.
2) On the DIV 6 a connectivity optimization process starts, characterized by connection pruning and unification of parallels connections.
3) The last process is migration. It occurs along the entire life cycle, and results in a massive clusterization at the last days (DIV 9-10).
Evolution of a neuronal culture into a clustered network. The left upper and lower frames represent the initial (day 0) and mature (day 12) configurations of the whole cultured area (size 7.7×6 mm2). Rectangles identify a specific area (size 1.6×1.8 mm2), whose intermediate evolution stages are reported in the other frames, ordered clockwise (see arrows). These latter snapshots correspond to days 0, 3, 5, 7, 10 and 12, respectively.
Another aspect of the neuronal network organization is the emergence of motifs and connectivity patterns during the growth of isolated or interacting clusters of neurons. Our hypothesis is that the development of motifs and circuits is an emergent phenomenon during the growth of the cultured network, and therefore, these motifs are formed in a much shorter temporal window than the time needed for reaching the network’s mature state. On its turn, this means that the evolution of the number of motifs per unit time will abruptly increase when the growth mechanism is guided by an optimization of the neuronal cable, while it will be constant in the case of an emergent phenomenon. We also plan to study the network morphology during the growth of an isolated cluster of neurons, as well as that one resulting from the interaction between two clusters of neurons, in order to elucidate the extent of complexity emerging from the interactions between the two modules.
Simultaneously with the laboratory work, we develop numerical models of the experimentally observed processes to gain a deeper insight in the study of the interplay between structure and dynamics in the cultured networks. We use biologically sounded models of neurons growing in spatial networks, in close equivalence with the experimental conditions. These models allow us to correlate the emergent dynamical properties, as synchronization or travelling waves, with the geometric properties of complex network, as clustering, centrality or modularity. Our results can be of relevance in all circumstances where one has access to functional measurements but the possibility of a full direct inspection of the topological structure is prevented as, for instance, functional brain measures.
Example of numerically generated network of 500 neurons showing the nodes with the highest topological (◦) and functional (•) centrality .
- Cognitive and Computational Neuroscience
- Computational Systems Biology
- Biological Networks
- Joint Italian-Spanish Laboratory on Complex Biological Networks
- Advanced Applied Mathematics to Biological Systems
Contact: Juan Antonio Almendral
 A. Rad, I. Sendiña-Nadal, D. Papo, M. Zanin, J.M. Buldú, F. del Pozo, S. Boccaletti, Phys. Rev. Lett. 108, 228701(2012).