Abstracts

Universal and non-universal scaling in brain activity: a group-normalization perspective.

Miguel A. Munoz

University of Granada, Spain

ABSTRACT: The human brain is in a state of perpetual reverberating activity, even in the absence of stimuli and tasks. Shedding light onto the origin and functional meaning of such an activity has become essential to understand how the brain represents, processes and stores information, i.e. on “how it works”. An inspiring but controversial conjecture proposes that operating in the vicinity of a critical point, with its concomitant power-laws and scaling, (or, similarly, “at the edge of chaos”) could provide brain networks with key advantages for information processing and justify many of their empirically-observed dynamical features.

In this talk, I will very briefly summarise the “criticality hypothesis” focusing on the non-equilibrium Landau-Ginzburg field theory developed by our group. Then, I will present new results from a recently-proposed phenomenological renormalization group (RG) approach that allowed us to analyze actual data for the activity of thousands of individually recorded neurons. These analyses lead to a number of remarkable non-trivial features, such as the existence of non-Gaussian fixed-point probability distributions and a robust set of critical exponents, which is rather universal across brain regions. I will scrutinize these results under the light of a very recent theoretical approach, which imposses a necessary condition for the neural representation of visual inputs to be robust (continuous and differentiable) in terms of the properties of the neural activity (co)variance. This study allows us to distinguish between universal, noisy, background activity and non-universal, input-driven activity. Remarkably, the covariance matrix eigenspectrum in both types of activities decays as a power law much as in critical phenomena. Finally, I will discuss results in a type of artificial neural network trained to perform an image-classification task, in which optimal results and experimental-like power-law decaying eigenspectra are found when the network is set to operate “at the edge of chaos”.

Estimating causal effects in Cell Transformation Assays

Federco M.Stefanini

University of Milan, Italy

ABSTRACT: Cell transformation assays (CTAs) are in vitro methods used in the preliminary

assessment of the carcinogenic potential of substances. Besides being quick-and-cheap, CTAs also enable the reduction of animal-based testing because experimental units are Petri dishes seeded with immortalized cells.

Despite CTAs are randomized one-way experiments where the experimental

factor is defined by 5 or more increasing concentrations,

different families of distributions have been proposed to evaluate the effect

of a substance on counts of Type III foci within Petri dishes.

Can the description of the data generating process be improved further by considering the total number of foci and the viability of treated cells, besides the number of fully transformed foci?

A parametric Bayesian structural causal model for BALB/c 3T3 CTAs has been recently proposed to distinguish total, (natural) direct, and indirect effects of a carcinogen. We argue that Natural Direct Effect (NDE) might be the right quantity to judge the carcinogenic potental of a substance.

The sample size required to reduce the expected uncertainty of estimated effects below a preassigned threshold may be calculated by Monte Carlo simulation after selecting the type of effect and the magnitude to detect.

Concluding remarks deal with limits and potential improvements of BALB CTA.

Bayesian Networks and Their Extensions in Modern Machine Learning

Marco Scutari

Istituto Dalle Molle di Studi sull'Intelligenza Artificiale (IDSIA), Switzerland

ABSTRACT: Bayesian networks are a key model in machine learning because of their flexibility and their intuitive graphical representation. In this seminar I will discuss how Bayesian networks have been adapted to address a number of different problems including incomplete data, time series, and causal modelling.

The role of structural biology in biological sciences: case studies of microcrystalline proteins characterised via X-ray powder diffraction methods

Irene Margiolaki

University of Patras, Greece

ABSTRACT: Structural biology is the study of the molecular structure and dynamics of biological macromolecules, particularly proteins and nucleic acids. Knowledge of 3D structures of biological molecules plays a major role in both understanding important processes of life and developing pharmaceuticals. Among several methods available for structure determination, macromolecular X-ray powder diffraction (XRPD) has transformed over the past decade from an impossible dream to a respectable method. XRPD can be employed in biosciences for various purposes such as observing phase transitions, characterizing bulk pharmaceuticals, determining structures via the molecular replacement method, detecting ligands in protein-ligand complexes, as well as in situ detection of novel protein crystal forms upon controlled relative humidity variation using laboratory XRPD. This presentation aims to provide necessary elements of theory and current methods, along with highlighted case studies. We will demonstrate the value of in-house and synchrotron XRPD as an analytical tool in industrial protein-based drug screening, and its potential to help troubleshooting the production process and to provide information for further refining the manufacturing of pharmaceuticals. Selected examples will be presented regarding studies of pharmaceutical proteins and their complexes with organic ligands including Human Insulin as well as peptide drugs.

Functional Diversification and Exaptation: The Emergence of New Drug Uses in the Pharma Industry

Gino Cattani

Stern School of Business, New York University

ABSTRACT: The process by which new uses or functions for an existing artefact (e.g., technology) emerge is an important yet understudied source of innovation. We call this process functional diversification (FD). We measure and characterize FD by focusing on the emergent uses of a class of technological resources: pharmaceutical drugs. We show that FD contains both an adaptive and an exaptive part and that it exhibits several empirical regularities. We argue (or suggest) that FD is at the core of several theoretical perspectives – exaptation, Penrose’s firm growth and shadow option theories – whose complementarities could be integrated into a general framework to understand and manage the emergence of new uses from existing resources.

The long way towards creating the Brain's Super Stem Cell: from cell culture assays and transplantation experiments, to comparative neuroanatomy and cell reprogramming

Ilias Kazanis

University of Patras, Greece

ABSTRACT: The postnatal mammalian brain (including the human) retains low levels of cytogenic plasticity driven by different pools of neural stem and progenitor cells. Multipotent (neurogenic and gliogenic) neural stem cells (NSCs) are localized within specialized microenvironments (called niches) at the walls of the lateral ventricles and at the hippocampus, contributing to odor recognition and memory, respectively. However, their numbers decline significantly with aging. Unipotent, oligodendrogenic progenitor cells (OPCs) are dispersed throughout the brain, supporting homeostatic and post-injury myelin turnover, exhibiting high levels of age resistance. In addition, latent, multipotent NSCs (lNSCs) have been shown to become transiently activated in response to injury and to contribute to the regeneration of the brain tissue. In light of the fact that the mammalian brain’s capacity for regeneration is very poor, the ability to harness the best properties of each of the above-described stem/progenitor populations (multipotency, age resistance, activity in the parenchyma) in order to create the Brain’s Super Stem Cell (BSSC) could be the way forward. For this target to be reached we must: a) describe in high detail the convergent and divergent properties of these progenitors. b) Find experimental protocols to accurately isolate these cells in order to investigate their behavior in vitro. c) Establish simple cell assays enabling the manipulation of their properties in vitro. The ultimate aim would be to use information generated by all the above in order to reprogram cells of the brain into BSSCs. Examples of recent experimental work, falling within this line of work, will be presented and discussed.

Talent or Luck? The role of chance in everyday life and in scientific careers

Alessandro Pluchino

University of Catania, Italy

ABSTRACT: It is a common belief that individual success is mainly due to personal qualities such as talent, intelligence, skills, smartness, efforts, hard work or risk taking. And even if we are willing to admit that a certain degree of luck could also play a role in reaching significant achievements, it is rather common to underestimate the importance of external forces in individual successful stories. It is clear that these assumptions are, on one hand, at the basis of the current, celebrated meritocratic paradigm, which affects not only the way our society grants work opportunities, fame and honors, but also the strategies adopted by Governments in assigning resources and funds to those who are considered the most deserving individuals. On the other hand, they appear to be in direct contrast with several studies which found that chance seems decisive in many aspect of our life [1]. A macroscopic quantitative indication questioning the exclusive role of talent in reaching success can be also found by comparing the statistical distributions of these two observables among a population. Actually, while talent – as the other human qualities, like for example intelligence – exhibits the typical Gaussian distribution [2], the distribution of wealth – often considered a quantitative proxy of success – is very asymmetric, with a large majority of poor people and a very small number of billionaires. In particular, as originally discovered by Pareto [3], the wealth distribution (of a country, for example) shows a power law tail, where the 20% richest individuals own 80% of the total capital and the 80% own the remaining 20%. Such a large discrepancy suggests that something beyond talent has to be at work behind the scenes of success. In a first study [1], recently awarded with the Ig Nobel prize for Economics, we supported the hypothesis that such an ingredient could be just randomness, amplified by the complex positive feedback mechanisms present in our socio-economic system. With the help of a simple agent-based model, called TvL (Talent vs Luck) model, we showed that these two ingredients (randomness and the “rich get richer” mechanism) are able to spontaneously bringing out a heavy-tailed distribution of capital P(C) in a population of individuals in spite of the non-heavy-tailed distribution of talent P(T). Moreover, we showed that the maximum success never coincides with the maximum talent, and viceversa, concluding that a certain degree of talent is surely necessary, but not sufficient, to reach high levels of success: context and random opportunities are equally fundamental. In a later work [2], we approached the same model from an analytical point of view, showing that it is possible to find the exact mathematical form of P(C) (not necessarily a power-law) only assuming an homogeneous distribution of talent. A greater talent heterogeneity produces an increase in the complexity of the analytical approach, making the task of finding a formal analytical relationship between the distributions of capital, talent and luck in the TvL model really a very hard, and still open, problem. Finally, in a last paper [4], we applied the TvL model engine to evaluate the importance of chance in scientific careers. To this aim, we reproduced in a virtual environment the publication-citation dynamics of the physics research community, calibrating the control parameters through the information extracted from the American Physical Society (APS) data set. Our results corroborate the hypothesis that also in the scientific context individual talent is necessary but not sufficient to reach high levels of success, since external factors could still play a fundamental role. These findings further question the naively meritocratic assumptions still surviving in the context of scientific research and explain why the strategies assigning honors, funds or rewards based only on individual past performances often fails their objectives [1].

[1] A. Pluchino, A. E. Biondo, A. Rapisarda, Talent versus luck: The role of randomness in success and failure, Advances in Complex Systems 21(3,4), 1850014 (2018). Awarded with the "Ig-Nobel Prize 2022 for Economics" on 15/09/2022.

For more info see: http://www.pluchino.it/talent-vs-luck-ita.html; http://www.andrea-rapisarda.it/ig-nobel-2022

[2] D. Challet, A. Pluchino, A. E. Biondo, A. Rapisarda, The origins of extreme wealth inequality in the Talent versus Luck model, Advances in Complex Systems 23(2), 2050004 (2020).

[3] V. Pareto, Cours d’Economique Politique, vol. 2 (1897).

[4] A. Pluchino, G. Burgio, A. Rapisarda, A. E. Biondo, A. Pulvirenti, A. Ferro, T. Giorgino, Exploring the Role of Interdisciplinarity in Physics: Success, Talent and Luck, PLoS ONE 14(6): e0218793 (2019).

Network physiology of cortico–muscular interactions: reorganization with sleep stages transitions and neurodegenerative disorders

Rossella Rizzo

University of Palermo, Italy

ABSTRACT: The brain plays a central role in regulating physiological and organ systems, including the skeleto-muscular and locomotor system. However, the brain-muscle communication network remains not understood. Traditional approaches to cortico-muscular coordination focus on associations between movement tasks or exercises and the activation of particular brain waves at specific cortical areas. However, neural control of the muscular system is continuously present even at rest. Moreover, the possibility to treat movement disorders has recently met with considerable interest in medical research. Specifically, a deep understanding of Parkinson's Disease (PD), the second most common progressive neurodegenerative disorder affecting older adults, is strictly correlated to a deep comprehension of brain control on locomotor system. Besides, changes in sleep regulation and rapid-eye-movement (REM) sleep behavior disorder (RBD) can appear in the early stage of disease and even prior to the onset of motor symptoms, and can serve, then, as a biomarker for PD. We hypothesize that network interactions between brain waves and rhythms embedded in muscle activity may also reflect changes in physiologic regulation as a function of physiological states. Further, we study the reorganization of brain-muscle networks with PD across sleep stages, developing useful biomarkers and providing a deeper understanding on the impact of PD on human organism networks.

We investigate the coupling between physiologically relevant brain waves at distinct cortical locations with peripheral EMG activity in different frequency bands across four major, well defined physiological states — Wake, REM, Light Sleep (LS), Deep Sleep (DS). Particularly, in the first part we analyze cortical EEG signals and surface chin and leg muscle tone EMG signals from 36 healthy young subjects; secondly, we consider data from 97 healthy subjects and 33 PD sub age matched. Utilizing a novel approach based on the Network Physiology framework and the concept of time delay stability (TDS) we find that for each physiologic state the network of cortico-muscular interactions is characterized by a specific hierarchical organization of network links strength, where particular brain waves are main mediators of interaction and control of muscular activity rhythms.

We discover a hierarchical reorganization in network structure across physiologic states, with high connectivity and network link strength during wake, intermediate during REM and LS, and low during DS, a sleep-stage stratification that demonstrates a unique association between physiologic states and cortico-muscular network structure. However, we notice some differences between healthy older subjects and PD subjects: the formers have stronger network connectivity during wake and LS, while the latter manifest a gradual decline in link strength from wake to DS. Indeed, the network connectivity and link strength are lower for PD than for healthy subjects during all sleep stages, but REM, when the PD subjects show a stronger brain-muscle interaction. Moreover, within each sleep stage, also the profile of network links strength as function of cortical rhythms frequency breaks down with PD, showing different relations between brain waves in link strength.

Our findings demonstrate previously unrecognized basic principles of brain-muscle communication, network integration and control, with potential clinical implications for neurodegenerative, movement and sleep disorders, and for developing efficient treatment strategies. Further, our studies could finally shed light on the connection between sleep behavior disorder and PD, and allow PD diagnosis in early course of the disease.