Workshop

MORNING SYMPOSIUM. Applied And Functional Modelling: Fmri, Graph Theory, Functional Connectivity, Analysis And Interpretation Of Cognitive Data

8:45–9:00 Welcome and Registration.

9:00–9:45 Multiscale functional and structural brain networks in health and disease

Jesus Cortes, BioCruces Hospital Clinic, Bilbao, Spain. 

9:45–10:30 Discovering functional networks using coupled cortical oscillations

Craig Richter, Basque Center On Cognition, Brain and Language (BCBL), San-Sebastian, Spain.

10:30–11:00 COFFE BREAK 

11:00–11:45 Pathotopologies’ dynamics and astrocytes: Evidence from animal models

Paolo Bonifazi, Bio Cruces Hospital Clinic, Bilbao, Spain. 

11:45–12:30 Neuron-Glial circuits for cognition.

Maurizio De Pitta, Basque Center for Applied Mathematics (BCAM), Bilbao, Spain, and the Krembil Research Institute, Toronto, Canada.

12:30–14:00 LUNCH

AFTERNOON TOPIC SESSION. Modeling Neuron-Glial Interactions

Curator: Maurizio De Pitta (BCAM/Krembil)

14:00–16:00 From Tripartite-Synapses to Neuron-Glial Networks

16:00–16:30 BREAK

16:30–18:00 Focus Session: Bistable Tripartite Synapses Synapses

MORNING SYMPOSIUM. Fundamental Models in Computational Neuroscience: McCulloch-Pitts Neurons and Hopfield Networks, Spin Glass Thermodynamics, Attractor Dynamics, and Chaos.

8:45–9:00 Setup

9:00–9:45 Nonequilibrium associative memories in neuroscience and machine learning Setup

Miguel Aguilera, Basque Center for Applied Mathematics (BCAM), Bilbao, Spain

9:45–10:30 Feedback Ising models neural networks dynamics

Daniele De Martino, Biofizika Institute, Leioa, Spain

10:30–11:00 COFFE BREAK 

11:00–11:45 Nonlinear dynamics in small groups of Hindmarsh-Rose neurons 

Dmitry Sinelshchikov, Biofizika Institute (CSIC-UPV/EHU), Leioa, Spain

11:45–12:30 Emergence of modular and hierarchical neural networks driven by learning of external stimuli

Gorka Zamora-Lopez, Pompeu Fabra University, (UPF), Barcelona, Spain 

12:30–14:00 LUNCH

AFTERNOON TOPIC SESSION. Glial Functional Readouts.

Curator: Maurizio De Pitta (BCAM/Krembil)

14:00–16:00 Hodgkin-Huxley models for Glial Calcium 

16:00–16:30 BREAK

16:30–18:00 Focus Session: Attractor dynamics in a GPCR model of astrocyte readout.

MORNING SYMPOSIUM. Brain-Inspired Modeling Tools: TVB and Neuro-AI approaches.

8:45–9:00 Setup

9:00–9:45 BiaPy: Bioimage analysis workflow for all audiences 

Ignacio Arganda-Carreras, University of the Basque Country (UPV/EHU), DIPC, Biofisika Institute, Leioa, Spain 

9:45–10:30 Digital twin technology to better understand and treat the brain 

Christophe Bernard, Inserm and Aix-Marseille University (AMU), Marseilles, France 

10:30–11:00 COFFE BREAK 

11:00–11:45 The Virtual Brain (TVB) – Architectural view over a modeling tool for full brain simulation

Romina Baila, CODEMART, Cluj-Napoca, Romania 

11:45–12:30 Focus Session: How to Harness The Virtual Brain Capability.

Romina Baila, CODEMART, Cluj-Napoca, Romania

12:30–14:00 LUNCH

AFTERNOON TOPIC SESSION. Physics of Neuron-Glia-Vascular Interactions.

Curator: Michelangelo Volpi (BCAM)

14:00–16:00 Conceptualization of the Glymphatic System and the Physics behind it.

16:00–16:30 BREAK

16:30–18:00 Focus Session: Computational Fluid Dynamics in the NGVU.

Ikerbasque Research Fellow, Caixa Junior Leader Fellow at  Basque Center for Applied Mathematics, Bilbao, Spain

WEBSITE: https://maguilera.net

Miguel Aguilera is an Ikerbasque Research Fellow and a la Caixa Junior Leader Fellow at BCAM – Basque Center for Applied Mathematics. He uses methods from complex systems research and related areas (nonequilibrium statistical mechanics, information theory, machine learning, and nonlinear dynamics) to study the principles of neural information processing and adaptive behaviour for systems in closed-loop interaction with their environments. 

Nonequilibrium associative memories in neuroscience and machine learning

Hopfield networks are a foundational model exemplifying how simple networks can exhibit complex behaviors reminiscent of memory storage and retrieval. Hopfield networks as a conceptual framework has profoundly impacted several disciplines, spanning neuroscience, statistical physics, and machine learning. More recently, an increasing interest in Hopfield networks as associative memories has surged within the machine learning community, due to the correspondence of Hopfield networks with attention mechanisms in transformer models, offering new avenues for exploring theoretical principles and designing innovative architectures. While Hopfield networks have traditionally been studied through the lens of equilibrium statistical physics, neural computation operates (in a thermodynamic sense) as an out-of-equilibrium, non-stationary process that changes dynamically, giving rise to entropy dissipation. In this talk, we will explore into the implications of extending Hopfield networks to the framework of nonequilibrium statistical physics, unveiling a diverse array of dynamic phenomena. This extended framework leads to applications in various domains, ranging from emergence of temporal asymmetries in neural circuits to applications to AI models like transformers employed in large language models.

Ikerbasque Research Associate, University of the Basque Country (UPV/EHU), Biofisika Institute

Research Associate, Donostia International Physics Center (DIPC)

WEBSITE_1: https://www.ikerbasque.net/en/ignacio-arganda-carreras

WEBSITE_2: https://www.biofisika.org/en/research/bioimage-analysis-and-computer-visi

Ignacio Arganda-Carreras is an Ikerbasque Research Associate at the Department of Computer Science and Artificial Intelligence of the University of the Basque Country (UPV/EHU) and also an affiliate of the Donostia International Physics Center (DIPC). He is one of the original developers of Fiji, one of the most popular open-source image processing software in the world, and widely used by the bio-image analysis community. His lab is focused on image processing and machine learning, especially in developing open-source computer vision methods for biomedical images.
 

BiaPy: Empowering Bioimage Analysis Workflows for All Audiences

BiaPy stands as a versatile and open-source bioimage analysis library designed to meet the diverse needs of users and developers in the field of computational neuroscience. With an intuitive interface, zero-code notebooks, and Docker integration, BiaPy offers accessibility across a wide spectrum of technical expertise. This presentation will delve into the core functionalities of BiaPy, highlighting its capacity to address bioimage analysis challenges at different scales. From beginners to experts, BiaPy provides customizable solutions for a range of tasks, including semantic segmentation, instance segmentation, object detection, and more. Through its robust support for multi-GPU configurations and compatibility with large datasets, BiaPy aims to democratize the use of sophisticated bioimage analysis workflows. Join us to explore how BiaPy empowers researchers to harness the power of deep learning and advanced computational methods in their work, facilitating breakthroughs in computational neuroscience.

Software engineer, CODEMART, Cluj-Napoca, Romania

WEBSITE: www.codemart.ro

Romina Baila is a software engineer based in Cluj-Napoca, Romania. She has been part of the TVB development team for the past 3 years, working on the brain simulator and making it more accessible for the researchers through the EBRAINS initiative. She has a Bachelor’s and a Master’s degree from the Technical University of Cluj-Napoca, with both her theses focusing on NLP applications.

The Virtual Brain (TVB) - Architectural view over a modeling tool for full brain simulation. Focus session: How to Harness The Virtual Brain Capability.

The Virtual Brain (TVB) represents a framework for computing, simulating, and analyzing the human brain, providing innovative insights about its structural and functional components. Simulating our brains is no trivial task, but the TVB team has spent well over a decade researching this topic, aiming to help the neuroscience community in understanding and better treating complex neurological conditions, like epilepsy. Over the past years, the TVB project has grown into an entire ecosystem, consisting of multiple connected modules, which aim to facilitate the understanding of how our brain works for researchers, doctors and patients. In this presentation, we will delve into what it means to build a human brain simulator and we will take a look at the modules that constitute this tool. In the interactive focus session everyone will get to run, analyze and visualize their own brain simulations, getting a hands-on feeling of why TVB is such a powerful application in the neuroscience field.

Director of Research, Institute of Systems Neuroscience, Inserm and Aix-Marseille University, Marseille, France

Editor in Chief of eNeuro

WEBSITE: https://ins-amu.fr/physionet

Christophe Bernard is Director of Research at the Institute of Systems Neuroscience, Inserm U1106, in Marseilles, France. He obtained a Ph.D. in Neuroscience from Paris VI university, did his postdoc at Southampton university, and a sabbatical In Houston (Baylor). His main interest is to understand brain dynamics in health and disease, with a focus on epilepsy (Michael Prize). He participated in the development of organic technologies to record and control brain activity (Felix Innovation Prize) and in The Virtual Mouse Brain (a platform that allows the virtualization of individual mouse brain to study whole brain dynamics in silico). He acts as a reviewing editor for Science Advances, and formerly for Science and Journal of Neuroscience. He created and is the Editor in Chief of eNeuro, the online open access journal of the Society for Neuroscience. eNeuro is designed to serve and educate the community, promoting reproducibility, publishing negative results, and sensitizing scientists to open science and better data interpretation with a focus on statistics and experimental bias.

Digital twin technology to better understand and treat the brain

Each brain is unique. A personalised approach is thus necessary to treat neurological disorders. This approach needs to be at the organ scale as all brain regions interact with each other in a complex manner. Whole brain modelling allows investigating mechanisms in individuals. Such models rely on identifying the rough anatomy (the regions) and how regions are connected to each other (the connectome), information that is provided by MRI. I will show how individual human and rodent brains can be virtualised, and the type of information that can be extracted. I will provide a concrete example of a successful use of digital twin technology to improve neurosurgery outcome in epilepsy.

Ikerbasque Research Associate, Biobizkaia Health Research Institute, Hospital Universitario Cruces, Barakaldo, Spain

WEBSITE: https://www.ikerbasque.net/paolo-bonifazi

Paolo Bonifazi is an Ikerbasque Professor at the BioCruces Health Clinic. He received a Master's in Physics from the University of Perugia (Italy) and a Ph.D. in Neuroscience from SISSA (Trieste, Italy). His research spans from fundamental to clinical neuroscience, emphasizing information processing and pathology in connection with circuit topology. Paolo combines in-vitro and experimental approaches, calcium imaging, multi-electrode recordings, patterned optogenetics, and immunochemistry, with computational and data science studies rooted in complex networks theory, information theory, and machine-learning classification analysis. He pioneered several mainstream notions in current neuroscience research, such as (i) the first evidence of GABAergic hub neurons (Science 2009); (ii) probing how astrocyte replenishment can restore connectivity and synchronization in dysfunctional networks (PNAS 2018), and (iii) demonstrating how acute inhibition of neuroinflammatory response via STAT3 pathway during epileptogenesis prevents GABAergic cells’ loss, reactive gliosis and imprinting of epileptic state (Brain 2023).

Linking hubs, embryonic neurogenesis, transcriptomics and diseases in human brain networks

In this talk I provide the basis for a multi-scale model of brain networks based on the neurogenesis time of the cerebral nodes, according to two major principles: “older gets richer” and “age preferential attachment”, both experimentally validated and inspired to the pioneering model on complex networks from Barabasi and Albert. This model bridges adult healthy human brain networks, to embryogenesis, and transcriptomics and its conclusions have been validated on the genetics underlying major brain pathologies including epilepsy. The link of the genetics underlying epilepsy (via STAT4) and the result of my last work (BRAIN, 2023) on prevention of epileptogenesis (via STAT3) will be presented. 

Full Professor at Ikerbasque, Computational Neuroimaging Lab, BioBizkaia Health Research Institute, Hospital Universitario Cruces, Barakaldo, Spain

WEBSITE: https://www.jesuscortes.info

Dr. Jesus M. Cortes, Full Professor at Ikerbasque and head of the Computational Neuroimaging Laboratory at the BioBizkaia Health Research Institute, has  an interdisciplinary background in the field of neuroscience, integrating brain connectivity, neuroimaging, clinical and neurophysiological data, and advanced techniques in machine learning and Artificial Intelligence. Dr. Cortes has secured funding exceeding 8.10 million EUR and has led 13 of the 35 research projects in which it has been involved. Dr. Cortes has published 115 scientific articles, 90% of which are ranked in the first quartile of the Journal Citation Reports (JCR Q1). He has supervised over 30 postgraduate students and delivered 102 scientific talks. His commitment to excellence and professional development is internationally recognized, with teaching contributions at 24 universities across six different countries. Awarded for his academic excellence and interdisciplinary approach, Dr. Cortes has received distinctions such as the "Premio Extraordinario de Doctorado" , Fulbright Fellow, EPSRC Fellow, and Ramón y Cajal Researcher. His recent appointment as Director of R+D in the cognitive stimulation platform NeuronUP has enabled the development of pioneering data-driven strategies for personalizing and optimizing neurocognitive interventions.

Multiscale Structural and Functional Brain Networks in Health and Disease

The precise relationship between brain structural and functional connectivities has puzzled leading researchers in network neuroscience. Despite numerous efforts,  an accurate prediction of the structural connectivity from its functional counterpart remains a distant goal.  This talk will focus on functional and structural networks derived from MRI, obtained through dynamics similarity in BOLD time series and white-matter tracts. Beyond comparing networks at the link level, a structure-function modular-coupling has demonstrated mutual benefits in comprehending brain structure and function. Our laboratory has made significant strides in this field, including the development of a brain partitioning technique that identifies network modules relevant to both structure and function [1,2]. We have applied this methodology to healthy individuals [3-8] and clinical populations, such as Alzheimer's patients [9], traumatic brain injury cases [10], and individuals with autism [11]. This talk will  provide an overview of these studies, showcasing our multidisciplinary approach that integrates physics, complex networks, machine learning, and high-order interactions in brain networks. 

Ikerbasque Research Fellow, Biofisika Institute (CSIC-UPV/EHU), Leioa, Spain

WEBSITE: https://www.ikerbasque.net/en/daniele-de-martino. 

Daniele De Martino is an Ikerbasque research fellow at the Biofisika Institute (Leioa) where he is co-leading the theoretical biophysics group. His research interest focuses on the application of statistical physics methods to study quantitatively cellular interactions, in particular in the context of metabolism. He took a PhD in Statistical Physics at SISSA/ISAS (Trieste), and he has been working as a post-doctoral researcher with the Chimera group at the University of Rome, IST Austria, and Jozef Stefan Institute (Slovenia).

Feedback Ising models neural networks dynamics

Spin-spiking models analyzed with statistical mechanics play an important role to bridge cellular and macroscopic scales for understanding biological network dynamics (Amit 1992) and the associated inverse and inference problems are used to resolve the real network topologies (Shneidemann et al 2006). One caveat of this approach is the simplifying equilibrium hypothesis, that is unable to reproduce stylized dynamical behaviors like oscillations. In this talk, I will discuss classical lattice models in presence of a feedback between order and control parameters. In presence of the latter, out-of-equilibrium phase transitions triggering collective self-oscillations takes over equilibrium critical points. I will focus  on the fully connected feedback Ising model and its capacity to quantitatively model statistical features brain waves, in particular the elusive coexistence of oscillations and avalanches. Feedback models inherit the analytical tractability of their equilibrium counterpart that makes it possible fast and reliable inference calculations. I will conclude discussing the extension of the feedback scheme to the Tricritical Ising and Potts models leading to higher order bifurcation points and chaos.

Scientist, Krembil Research Institute, University Health Network, Toronto, Canada 

Assistant Professor, Department of Physiology, Temerty Faculty of Medicine, University of Toronto, Toronto, Canada

Scientific Associate, Basque Center for Applied Mathematics, Bilbao, Spain

Adjunct Professor, Department of Neurosciences, Faculty of Medicine, University of the Basque Country, Leioa, Spain

WEBSITE: https://sites.google.com/site/mauriziodepitta/home?authuser=0

Dr. Maurizio De Pitta is a computational neuroscientist specializing in studying neuron-glial interactions. He is the principal investigator at the Neuron-Glial Interaction Lab at the Krembil Research Institute in Toronto (Canada) and holds an Assistant Professor position at the University of Toronto's Department of Physiology, Temerty Faculty of Medicine. He also has affiliations with the Basque Center for Applied Mathematics and the University of the Basque Country in Spain. Dr. De Pitta earned his Ph.D. in Computational Biology from Tel Aviv University and conducted postdoctoral research in Theoretical Neurosciences at The University of Chicago. His research interests range from models of neuron-glial physiology in health and disease in cognition and age-related dementia, using multiple quantitative approaches borrowed from genomics, physics, statistics, and computer science. Dr De Pitta is also a principal investigator in the European H2020 ASTROTECH Consortium to develop forefront glial—brain interfaces and co-founder of the Spanish Clinical System Neuroscience Network, CliSyNe.

Neuron-Glial Molecular Switches Across The Cognitive Spectrum

Healthy brain functions rely on the intricate interaction of neurons with glial cells. Among the latter, astrocytes are ubiquitous in our cortical circuits and can affect synaptic transmission on multiple time scales. On the short time scale, they are responsible, for example, for glutamate clearance, which is critical in setting the tone of neural activity. On a longer time scale, astrocytes operate as endocrine cells, modulating synaptic function by releasing common transmitter molecules. Although different in nature, both pathways may mediate positive feedback on neural activity, resulting in the emergence of multistability. In this scenario, the multiple activity states emerging from neuron-astrocyte interactions could account for various cognitive-related mechanisms in the healthy and diseased brain: from working-memory tasks to dementia-related neural correlates.

Staff Scientist, Ramón y Cajal Fellow, Basque Center on Cognition, Brain and Language (BCBL), San-Sebastian, Spain 

WEBSITE: https://www.bcbl.eu/en/conocenos/equipo/craig-richter

Dr. Craig Richter is a Canadian Neuroscientist who obtained his PhD at the Center for Complex Systems and Brain Sciences, Florida Atlantic University, in 2010. Since then, he has resided in Europe performing research at the Ernst Struegmann Institute in Cooperation with the Max Planck Society (ESI), Frankfurt, Germany, and the Cognitive and Computational Neuroscience Lab, Ecole Normale Superieure, Paris, France. He is currently a Ramon y Cajal Fellow with at the Basque Center on Cognition, Brain and Language (BCBL), San-Sebastian, Spain where he studies how cortical oscillations relate to perception and cognition.

Discovering functional networks using coupled cortical oscillations

Since the seminal work of Hans Berger, it has been understood that cortex often shifts between oscillatory regimes, that heavily correlate with cognitive state, and behavioural goals. In this talk, I will focus on modern theories and findings that investigate how these oscillatory states may modulate the passing and processing of information between cortical areas in a dynamic and adaptive fashion.

Ikerbasque Research Fellow, Biofisika Institute (CSIC-UPV/EHU), Leioa, Spain

WEBSITE: https://www.biofisika.org/en/about/people/dmitry-sinelshchikov

Dmitry Sinelshchikov is an Ikerbasque Research Fellow at the Biofisika Institute (CSIC-UPV/EHU) (IBF) working in dynamical systems and ordinary differential equations with applications to biophysics, physics and medicine. Dmitry obtained his PhD in Mathematics and Physics (2010) from Moscow Engineering Physics Institute (MEPhI). After that Dmitry consequently worked from 2010 to 2019 as an Assistant Professor, Senior Lecturer and Associate Professor at MEPhI. During this period, his research was focused on analytical theory of differential equations and dynamical systems applied to fluid dynamics. Dmitry’s works on integrability of Li\'enard-type equations received recognition from the Russian Academy of Sciences and he was awarded the gold medal for young scientists in 2017. In 2019 Dmitry joined the HSE University as an Associate Professor, where he pursued his independent research line in integrability of dynamical systems and applied nonlinear dynamics. Since 2022 Dmitry is working in Spain at the IBF. In 2023 Dmitry was awarded simultaneously the Beatriz Galindo and Ikerbasque Research Fellow grants, of which he accepted the latter to continue working in biophysical applications of dynamical systems in a direct collaboration with experimental groups at the IBF. 

Nonlinear dynamics in small groups of Hindmarsh-Rose neurons

In this talk we consider nonlinear dynamics in small groups of neurons that are described by the Hindmarsh-Rose model. This model aims to describe the bursting and spiking activity of membrane potential and its interactions with fast and slow ionic currents. While the Hindmarsh-Rose model for one neuron is well studied, the dynamics of coupled Hindmarsh-Rose neurons is still not completely understood. Here we study a model of three Hindmarsh – Rose neurons with directional electrical connections. We consider two fully-connected neurons that form a slave group which receives the signal from the master neuron via a directional coupling. We control the excitability of the neurons by setting the constant external currents. We study the possibility of excitation of the slave system in the stable resting state by the signal coming from the master neuron, to make it fire spikes/bursts tonically. We vary the coupling strength between the master and the slave systems as another control parameter. We calculate the borderlines of excitation by different types of signal in the control parameter space. We establish which of the resulting dynamical regimes are chaotic. We also demonstrate the possibility of excitation by a single burst or a spike in areas of control parameters, where the slave system is bistable. We calculate the borderlines of excitation by a single period of the excitatory signal.

PhD Researcher, University of the Basque Country (UPV/EHU) and Basque Center for Applied Mathematics.

WEBSITE: https://michelangelov.github.io/michelangelovolpi-website/

Michelangelo Volpi is a Ph.D. researcher of the ASTROTECH Consortium, part of the Marie Skłodowska-Curie Innovative Training Network, at the Basque Center for Applied Mathematics (BCAM) and the University of the Basque Country (UPV/EHU).  His research focuses on the biophysical modeling of neuron-glial circuits, with an emphasis on vascular coupling. He holds a Master's in Physics of Complex Systems from the University of Turin (Italy) and a Bachelor's in Physics from the University of Milan-Bicocca (Italy).

Conceptualization of the Glymphatic System and the Physics behind it 

Recent research has uncovered the existence of a clearance mechanism in the central nervous system: the glymphatic system. It works as a substitute for the lymphatic system and comprises neuronal structures, glial cells, and blood vessels, all regulated by astrocytes in specialized functional territories that may range in size from a few nanometres up to many centimeters. The underpinning functioning involved water fluxes facilitating the movement of cerebrospinal fluid, interstitial fluids, and metabolic waste from the arterial perivascular space to the venous circulation. In this physiological framework, all the electrical signaling processes, which involve ion exchanges between intra and extracellular compartments, are affected by fluid transport. In this direction, this talk will focus on the biophysical description of the glymphatic pathway, exploring the mathematical foundations of transport equations and limitations in modeling.

Senior research assistant, Center for Brain and Cognition, Dept. of Information and Communication Technologies, Pompeu Fabra University

Post-doctoral Research Assistant and Project Manager, The Human Brain Project

WEBSITE: www.Zamora-Lopez.xyz

Gorka Zamora-López studied theoretical physics at the University of the Basque Country (UPV/EHU), Spain, and biophysics at the University of Oulu, Finland. During his Ph.D. at the University of Potsdam he began an interdisciplinary journey at the cross-road between complex networks, neuroscience and dynamical systems, discovering the rich-club organization of brain connectomes in early maps of white-matter. Since 2013 he is a reasearcher at the Center for Brain and Cognition (CBC) of the Pompeu Fabra University, solving several methological matters for (brain) connectivity, the understanding of the structure-function relation in the brain's organization and delving into clinical research such as disorders of consciousness. As part of the Human Brain Project, for seven years, he combined the daily research with the coordination of technological solutions for scientific computational pipelines.

Emergence of modular and hierarchical neural networks driven by learning of external stimuli

In the last three decades the field of brain connectivity has explored the function of the white matter. Beyond the specialization of individual cortical regions, we have found that the cortex is organised into a modular and hierarchical architecture that supports the coexistence of segregation and integration of information. A prominent remaining question is to understand how the brain could possibly evolve into such a network. Here, we give a first step into answering this question and propose that adaptation to various inputs could have been a relevant driving mechanism. To illustrate that, we develop a model of (quadratic integrate and fire) spiking neurons, subjected to stimuli focalised on different populations. We find that only the combination of Hebbian and anti-Hebbian inhibitory plasticity allows the formation of stable modular organization in the network. To add further biological plausibility, our model continues "alive" after the entrainment, setteled into an aysnchronous irregular dynamics. We find that the emergence of spontaneous memory recalls during this resting-state activity is crucial for the long-term consolidation of the learned memories.