Cicle de Webinars sobre Sistemes Complexos enfocats al COVID-19 organitzat per l'UBICS i complexitat.cat. Dirigit a tothom interessat en els Sistemes Complexos i els seus actuals camps d’aplicació.
11 Juny (16h) - Manlio de Domenico, CoMuNe Lab, Fondazione Bruno Kessler
"Tackling complexity: foundations and appplications"
Abstract: Complex systems consists of units whose interactions at a microscopic scale lead to the spontaneous emergence of collective behavior and other unexpected phenomena at the meso- and macroscale. In this seminar I will introduce some basic concepts and tools of complexity science without relying on technicalities. In the second part of the seminar I will briefly discuss the relevance of big data for the analysis of complex systems and, more specifically, of socio-technical systems, spanning from the rise of collective attention to one of the most relevant phenomena observed during the COVID-19 pandemic: the infodemic related to coronavirus.
18 Juny (16h) - Nuria Oliver, Data-Pop Alliance & ELLIS (The European Laboratory for Learning and Intelligent Systems)
"Data Science to fight against COVID-19"
Abstract: In my talk, I will describe the work that we have done within the Commission on AI and COVID-19 for the President of the Valencian Region. As commissioner, I have led a multi-disciplinary team of 20+ scientists who have volunteered since March 2020. We have been working on 4 large areas: (1) human mobility modeling; (2) computational epidemiological models (both metapopulation and individual models); (3) predictive models; (4) citizen surveys: https://covid19impactsurvey.org.
I will describe the results that we have produced in each of these areas and will share the lessons learned in this very special initiative of collaboration between the civil society at large (through the survey), the scientific community (through the Expert Group) and a public administration (through the Commissioner at the Presidency level).
25 Juny (16h) - Santiago F. Elena, Instituto de Biología Integrativa de Sistemas, CSIC
“Identifying early-warning signals for the sudden transition from health to disease stages by dynamical network biomarkers”
Abstract: One of the most outstanding observations during COVID-19 pandemics is that some patients have an asymptomatic infection while others suffer severe symptoms, some of them becoming fatal. Is there any relation between the global gene expression state of patients and they propensity to suffer an asymptomatic infection? Is it possible to identify which genes, or groups of them within a regulation network, may serve as markers to predict the clinical fate of a patient before the presence of any symptom? In this seminar I will introduce the fundamentals of the theory of Dynamical Biomarkers of Networks (DBN), illustrating their application to the analysis of transcriptomic data during disease progression in a pathosystem model.
02 Juliol (16h) - Alex Arenas, DEIM, Universitat Rovira i Virgili
“Epidemics and mobility”
Abstract: Reaction–diffusion processes have been widely used to study epidemics in networked metapopulations. In the context of epidemics, reaction processes are understood as contagions within each subpopulation (patch), while diffusion represents the mobility of individuals between patches. Recently, the characteristics of human mobility, such as its recurrent nature, have been proven crucial to understand the phase transition to endemic epidemic states. Here, we present a framework able to cope with the elementary epidemic processes, the spatial distribution of populations and the commuting mobility patterns. We will show after, how this framework has been adapted to describe the COVID-19 pandemic.
16 Juliol (16h) - Ernesto Estrada, IUMA
“Fractional difusion on the human proteome as an alternative explanation to the multi-organ damage of SARS-CoV-2”
Abstract: SARS CoV-2 is the new coronavirus causing the pandemic known as COVID-19. This respiratory disease is characterized by multi-organ and systemic damages in patients. The abundance of ACE2 on human organs has been claimed as responsible for such multi-organ spread of the virus damages. However, once on circulation the virus could spread to practically every organ in the human body as ACE2 is ubiquitous on endothelia and smooth muscle cells of virtually all organs. Contrastingly, SARS CoV-2 only damages selectively a few organs. Here, we develop the hypothesis that the effects of the SARS CoV-2 virus can be spread through the human protein-protein interaction (PPI) network in a subdiffusive way. We then elaborate a time-fractional diffusion model on networks which allow us to study this phenomenon. Starting the diffusion from the SARS CoV-2 Spike protein to the human PPI network we show here that the pertubations can spread across the whole network in very few steps. Consequently, we discover a few potential routes of propagation of these perturbations from proteins mainly expressed in the lungs to proteins mainly expressed in other different tissues, such as the heart, cerebral cortex, thymus, lymph node, testis, prostate, liver, small intestine, duodenum, kidney, among others already reported as damaged by COVID-19.
Aneta Stefanovska (Lancaster University, UK)
In the real world, any system under study is subject to external perturbations. Mostly, these are either neglected, taken as part of the system, or treated as a noise. In this talk, we propose a fourth approach, which is to treat the system under study as being nonautonomous. We consider the particular case where there are perturbations to the phase of an oscillatory system. We will discuss the stability properties of such coupled systems and networks, and present a set of algorithms, from our toolbox MODA, that can be used for extracting their finite-time dynamical properties. The approach will be illustrated by applications to living systems including cells and the brain, as well as to physical systems such as rogue waves and electrons on the surface of superfluid helium.
Dr. Markus Abel (Potsdam University & Ambrosys, Germany)
Recently, the term explainable AI became known as an approach to produce models from artificial intelligence which allow interpretation. Since a long time, there are models of symbolic regression in use that are perfectly explainable and mathematically tractable: in this contribution we demonstrate how to use symbolic regression methods to infer the optimal control of a dynamical system given one or several optimization criteria, or cost functions. In previous publications, network control was achieved by automatized machine learning control using genetic programming. Here, we focus on the subsequent analysis of the analytical expressions which result from the machine learning. In particular, we use AUTO to analyze the stability properties of the controlled oscillator system which served as our model. As a result, we show that there is a considerable advantage of explainable models over less accessible neural networks.
Hugues Chaté (CEA-Saclay & Beijing CSRC, PRL Editoral Board)
Active matter consists of elementary units producing mechanical work to move themselves or to displace other objects. In other words, active matter is about systems maintained out-of-equilibrium "in the bulk", burning energy to produce directed, persistent motion.
This very general definition covers all kinds of situations at all scales: groups of animal or robots, collective of cells and micro-organisms, active colloids and phoretic swimmers, mixtures of biofilaments and motor proteins. Most active matter systems exhibit surprising if not spectacular emerging collective properties that we are only starting to understand.
In this talk, I will strive to give a synthetic and organized overview of what is still a fast-growing field. This overview will however be rather personal, drawing mostly from my own work. A large part if the talk will be devoted to presenting experimental results obtained mostly on living active matter, which should be of interest to biologists.
Esta jornada quiere ser una invitación al diálogo interdisciplinar y a la puesta en común de perspectivas metodológicas, visiones sociopolíticas, resultados concretos y líneas de investigación acerca de cómo los entornos digitales han cambiado la estructura de la comunicación y las gramáticas de interacción social. Un primer paso para constituir espacios en los que la ciencia de la comunicación, la sociología, la politología y la ciencia de los sistemas complejos, entre otras disciplinas, colaboren para mejorar nuestra comprensión de lo que ocurre en las plataformas sociales.
Heurística + Tecnopolítica