Complexity Science and the Brain
We are pleased to invite you to our upcoming scientific workshop, which aims to provide an informal interactive platform to discuss recent advances of computational techniques for analyzing brain dynamics. Approaches such as dynamical systems theory, information theory, dynamic functional connectivity, and graph theory have all proven to be useful tools in this field. However, there is currently no clear consensus on how to merge these tools or which tool is most appropriate for a given analysis.
The workshop will gather practitioners in the field that will present their work but also discuss strengths and limitations of current approaches and the challenges ahead. We believe this workshop will be a valuable opportunity for participants to exchange ideas, strengthen existing collaborative work and forge relationships.
Organizers
Federico E. Turkheimer, Professor in Neuroimaging, IOPPN, King's College London
Paul Expert, Lecturer in Health Informatics, UCL Global Business School for Health
Giuseppe de Alteriis, PhD Student, IOPPN, King's College London
Julia Schubert, Research Associate, IOPPN, King's College London
Laila Rida, PhD Student, IOPPN, King's College London
Benjamin Gooddy, PhD Student, IOPPN, King's College London
Bryony Goulding Mew, PhD Student, IOPPN, King's College London
Emilie Wielezynski, King's College London
Paul Expert, Lecturer in Health Informatics, UCL Global Business School for Health
Giuseppe de Alteriis, PhD Student, IOPPN, King's College London
Julia Schubert, Research Associate, IOPPN, King's College London
Laila Rida, PhD Student, IOPPN, King's College London
Benjamin Gooddy, PhD Student, IOPPN, King's College London
Bryony Goulding Mew, PhD Student, IOPPN, King's College London
Emilie Wielezynski, King's College London
The Speakers
Joana Cabral, University of Minho, Portugal
Functional brain networks and wave patterns: from function to generative mechanisms
Spontaneous fluctuations in functional magnetic resonance imaging (fMRI) signals correlate across distant brain areas, forming functional networks that appear disrupted in numerous psychiatric and neurological disorders, pointing to a key role in brain function. However, the generative mechanism of fMRI signal correlations are not fully understood. In this talk, I will give an overview of the evidence gathered applying Leading Eigenvector Dynamics Analysis (LEiDA) on different neuroimaging datasets and the insights it provides to understand brain function. I will further describe recent insights on the origin of LEiDA patterns obtained from experiments in rodent. Overall, despite the convincing evidence, it is crucial to obtain a better mechanistic understanding of functional networks and their organizing principles, in order to design informed strategies to rebalance long-range interactions between brain areas that appear disrupted in disease.
Joana Cabral is a Portuguese Biomedical Engineer. She did her PhD in Computational Neuroscience with Gustavo Deco, followed by a postdoc at the Psychiatry department in Oxford with Morten Kringelbach. She returned to Portugal in 2017, from where she mantains an extensive network of national and international collaborations. She has published 60 research papers covering a wide range of neuroimaging studies and modelling works. She has presented in over 40 international meetings. She was awarded the L’Oréal award for Women in Science in 2018 and is currently funded by Fundación La Caixa.
Giovanni Petri, Northeastern University London, UK
Higher-order fingerprinting: informational and topological signatures of individuality
Network neuroscience is a dominant paradigm for understanding brain function.Functional Connectivity (FC) encodes neuroimaging signals in terms of the pairwise correlation patterns of coactivations between brain regions. However, FC is by construction limited to such pairwise relations.Here, we explore functional activations as a topological space via tools from topological data analysis. In particular, we analyze the resting fMRI data of populations of healthy subjects across ages, and demonstrate that algebraic-topological features extracted from brain activity are effective for brain fingerprinting. By computing persistent homology and constructing topological scaffolds, we show that these features outperform FC in discriminating between individuals and ages.That is, the topological structures are more similar for the same individual across different recording sessions than across individuals. Similarly, we find that topological observables improve discrimination of individuals of different ages. Finally, we show that the regions highlighted by our topological methods are characterized by characteristic patterns of information redundancy and synergy which are not shared by regions that are topologically unimportant, hence establishing a first direct link between topology and information theory in neuroscience.
Giovanni Petri, PhD is a Professor in the Network Science Institute at Northeastern University London. He also holds affiliations as Principal Researcher at CENTAI, and as Guest Scholar in the Networks Units of IMT Lucca. Previously, he was Senior Research Scientist in the "Mathematics and Complex Systems" lab of ISI Foundation since 2016. He is a theoretical physicist that shortly after graduating decided that complex systems – in the broadest sense – were more intriguing than cosmology. He fell in love with the idea of high-order interactions, of emergent properties and ended up earning a PhD on complex networks at Imperial College London in 2012. Theoretical approaches never stopped fascinating him, and he continues this research today working at the interface between complex systems and algebraic topology. His research spans the analysis of neuroimaging data and AI systems with topological techniques, the formalization of cognitive control models with tools of statistical mechanics and network theory, and the study of the predictability of socio-technical systems.
Andrea Luppi, McGill University, Canada
Transitions between cognitive topographies: contributions of network structure, neuromodulation, and disease
To explain how the brain orchestrates information-processing for cognition, we must understand information itself. Importantly, information is not a monolithic entity. Information decomposition techniques provide a way to split information into its constituent elements: unique, redundant, and synergistic information. In this talk I will outline how disentangling synergistic and redundant interactions is redefining our understanding of integrative brain function and its neural organisation. To explain how the brain navigates the trade-offs between redundancy and synergy, I combine multimodal and multi-species evidence integrating the structural, molecular, and functional underpinnings of synergy and redundancy; their roles in cognition and computation; and how they might arise over evolution and development. Overall, disentangling synergistic and redundant information provides a guiding principle to understand the informational architecture of brain and cognition.
Andrea Luppi studied philosophy, neuroscience, and cognitive science at the University of Oxford and Harvard University, before obtaining a PhD in neuroscience from the University of Cambridge and the Alan Turing Institute. He is currently a Molson Neuro-Engineering Fellow and Neuro-AI Postdoctoral Scholar at the Montreal Neurological Institute of McGill University in Montreal, working in the group of Prof. Bratislav Misic. Andrea’s work investigates how cognition and consciousness arise from the complex interplay of network structure and brain dynamics across scales. To this end, he combines approaches from information theory, network science and whole-brain computational modelling to investigate the dynamics and connectivity of the brain across pharmacological and pathological perturbations. His long-term goal is to understand what makes artificial and biological neural networks capable of consciousness and intelligence.
Fernando Rosas, University of Sussex, UK
Formal approaches to emergence: theory, practice, and opportunities for brain analyses
Emergence is a profound subject that straddles many scientific scenarios and disciplines, including how galaxies are formed, how flocks and crowds behave, and how mental activity arises from the orchestrated activity of neurons. At the same time, emergence is a highly controversial topic, surrounded by long-standing debates and disagreements on how to best understand its nature and its role within science. This talk presents formal approaches to characterise emergence as way to advance these discussions. Specifically, we argue that these approaches make emergence useful for empirical practice, giving researchers rigorous frameworks to guide discussions and advance theories, and also quantitative tools to establish conjectures about emergence and then test them on empirical data. We provide an overview on the theory and practice of formal approaches to emergence, and highlights the opportunities they open for studying the relationship between the brain and the mind. The talk presents illustrative examples of the application of principles of emergence to practical data analysis, discussing several key interpretation issues and avenues for future work.
Fernando Rosas works as lecturer at the University of Sussex and research fellow at Imperial College London and the University of Oxford. His work is focused on the development of conceptual and computational tools to better understand the collective dynamics of complex systems. His background includes studies in mathematics, physics, philosophy, and music composition.
Huifang Wang, Aix-Marseille Université, France
Virtual brain twins in epilepsy and beyond
In this talk, we will have the definition of virtual brain twins, its key element and relationship between these elements. Then we will see the concept of personalized whole-brain network models. I will give a solid example in terms of the personalized whole-brain network modelling in epilepsy, which we call virtual epilepsy patients (VEP). I will explain how to use VEP to estimate epileptogenic networks, plan virtual surgeries and stimulations. In the end, we can extent this personalized whole-brain network modelling in other uses, such as healthy ageing, Alzheimer’s, multiple sclerosis, Parkinson’s and psychiatric disorders.
Dr. Huifang Wang is a neuroscientist who works with Dr. Viktor Jirsa in INS, an AMU and INSERM institute. Her current research is interested in personalized whole brain modelling (digital twins) in neurology. Led by Viktor, they have built a virtual epileptic patient pipeline for the diagnosis and treatment of epilepsy. Now we are on the journal to digital twins for the dignosis, treatement and prognosis for epilepsy and other brain disorders.
Henrik Jensen, Imperial College London, UK
Brain dynamics: Shadows of long-range correlations and criticality.
In statistical mechanics a system is said to be in a critical state if the correlation length is infinite. This implies the lack of specific characteristic scales, and the resulting self-similar structure of the system is characterised by power laws. The structure of brain dynamics shows many aspects of this type of critical behaviour. This has led some authors to look for ways to quantify “how critical” a particular brain is in a particular situation. We will mention why the breaking of scaling relations is an unlikely biometric measure of the degree of the brain’s criticality. We will suggest a better alternative is to focus on monitoring correlations in space and time combined with information theoretic measures.
Henrik Jensen is an expert on the statistical mechanics of complex systems. His most recent book is "Complexity Science: The Study of Emergence" (Cambridge Press) which presents complexity science as the organised scientific study of collective emergent behaviour. He has worked on the dynamical properties of condensed matter systems and developed the Tangled Nature model of evolving ecosystems. An approach of broad relevance to evolving adaptive complex systems. His work together with Paolo Sibani on record dynamics and its relevance to a surprisingly broad range of complex system, including macroevolution and the dynamics of ant colonies, have attracted interest from researchers from biology to materials science. Recently he has focused on brain dynamics and structure by analysing fMRI and EEG data making use of approaches inspired by statistical mechanics and information theory.
Fran Hancock, King's College London, UK
Metastability demystified – praeteritum, present, futurum
Healthy brain functioning depends on balancing stable integration between brain areas for effective coordinated functioning, with bursts of desynchronisation to allow subsystems to reconfigure and express functional specialisation. Metastability, a concept originated in statistical physics and dynamical systems theory, has been proposed as a key signature that characterises this balance. Building on this principle, the neuroscience literature has employed markers of metastability to investigate various aspects of brain function including cognitive performance, healthy ageing, meditation, sleep, responses to pharmacological challenges, and to characterise psychiatric conditions or disorders of consciousness. However, this body of work often uses the notion of metastability heuristically, and sometimes inaccurately, making it hard for the uninitiated to navigate the vast literature, interpret findings, and foster further development of theoretical and experimental methodologies. In this presentation I will provide an abridged review of metastability and its applications in neuroscience, recall its scientific and historical foundations, explore some practical estimators used in empirical data, and suggest ways forward for future developments.
Fran is a visiting research fellow at the IoPPN in King’s College London collaborating with Prof. Federico Turkheimer at the Centre for Neuroimaging Studies. Following a BSc in applied mathematics and physics, and an MBA, Fran worked for many years as a management consultant in the pharmaceutical industry before taking a senior executive role in the R&D organization of a global consumer goods company. She returned to academic pursuits completing a BSc in psychology, an MSc in Applied Neuroscience, and most recently, a PhD in applied neuroimaging at King’s. Given her diverse experience and academic interests, Fran strives for conceptual integration of concepts and theories that sometimes remain isolated in specific domains such as cognitive psychology, dynamical systems theory, information theory, or stochastic processes. One such concept is metastability for which Fran continues to investigate empirically, theoretically, and computationally.
Giuseppe de Alteriis, King's College London, UK
A Framework for brain-wide network dynamics: from calcium imaging to fMRI
Dynamic Functional Connectivity is a growing field of research in neuroimaging. In my talk, I will propose a new framework to calculate, analyze and interpret dynamic connectivity patterns based on their mathematical structure. I will explain why this framework has the level of abstraction necessary to analyze brain-wide network dynamics at different spatio-temporal scales, showing evidence from widefield calcium imaging and fMRI in rodents. Finally, I will show in different neuroimaging datasets how some of the brain-wide network properties are disrupted with ageing and in psychiatric disorders and how this is related to cognitive abilities.
Giuseppe is a LIDo (London Interdisciplinary Doctoral Programme) PhD student at King's College London and UCL. His research is focused on translating ideas from complex systems theory and network theory to the brain. His ambition is to work at different brain scales, from 2p imaging of single neurons in rodents to human fMRI, in order to investigate the neural basis of brain wide network dynamics.
Pedro Mediano, Imperial College London, UK
Entropy and Complexity in neural activity
There are many ways of measuring complexity in experimental data, each of them with different strengths and weaknesses. Here we will focus on the information-theoretic construct of entropy, which quantifies complexity as the degree of disorder in a system. We will cover several empirical results linking entropy in the brain with level of consciousness, and present new results exploring this relationship. Finally, we will finish by discussing new open-source tools available to estimate entropy from neural time series and show that these yield rich, meaningful analyses of experimental data.
Pedro holds a degree in Physics from the University of Valencia, a PhD at Imperial College and a postdoc at the University of Cambridge, where he worked on consciousness and complexity. Now, he is a lecturer at the Department of Computing, Imperial College London.
Broadly speaking, his research lies at the intersection between consciousness science, information theory, complexity science, and machine learning. He strives to ground his research in solid mathematical and theoretical grounds, and whenever possible he publishes open-source software to make his research accessible and reproducible.
Functional brain networks and wave patterns: from function to generative mechanisms
Spontaneous fluctuations in functional magnetic resonance imaging (fMRI) signals correlate across distant brain areas, forming functional networks that appear disrupted in numerous psychiatric and neurological disorders, pointing to a key role in brain function. However, the generative mechanism of fMRI signal correlations are not fully understood. In this talk, I will give an overview of the evidence gathered applying Leading Eigenvector Dynamics Analysis (LEiDA) on different neuroimaging datasets and the insights it provides to understand brain function. I will further describe recent insights on the origin of LEiDA patterns obtained from experiments in rodent. Overall, despite the convincing evidence, it is crucial to obtain a better mechanistic understanding of functional networks and their organizing principles, in order to design informed strategies to rebalance long-range interactions between brain areas that appear disrupted in disease.
Joana Cabral is a Portuguese Biomedical Engineer. She did her PhD in Computational Neuroscience with Gustavo Deco, followed by a postdoc at the Psychiatry department in Oxford with Morten Kringelbach. She returned to Portugal in 2017, from where she mantains an extensive network of national and international collaborations. She has published 60 research papers covering a wide range of neuroimaging studies and modelling works. She has presented in over 40 international meetings. She was awarded the L’Oréal award for Women in Science in 2018 and is currently funded by Fundación La Caixa.
Giovanni Petri, Northeastern University London, UK
Higher-order fingerprinting: informational and topological signatures of individuality
Network neuroscience is a dominant paradigm for understanding brain function.Functional Connectivity (FC) encodes neuroimaging signals in terms of the pairwise correlation patterns of coactivations between brain regions. However, FC is by construction limited to such pairwise relations.Here, we explore functional activations as a topological space via tools from topological data analysis. In particular, we analyze the resting fMRI data of populations of healthy subjects across ages, and demonstrate that algebraic-topological features extracted from brain activity are effective for brain fingerprinting. By computing persistent homology and constructing topological scaffolds, we show that these features outperform FC in discriminating between individuals and ages.That is, the topological structures are more similar for the same individual across different recording sessions than across individuals. Similarly, we find that topological observables improve discrimination of individuals of different ages. Finally, we show that the regions highlighted by our topological methods are characterized by characteristic patterns of information redundancy and synergy which are not shared by regions that are topologically unimportant, hence establishing a first direct link between topology and information theory in neuroscience.
Giovanni Petri, PhD is a Professor in the Network Science Institute at Northeastern University London. He also holds affiliations as Principal Researcher at CENTAI, and as Guest Scholar in the Networks Units of IMT Lucca. Previously, he was Senior Research Scientist in the "Mathematics and Complex Systems" lab of ISI Foundation since 2016. He is a theoretical physicist that shortly after graduating decided that complex systems – in the broadest sense – were more intriguing than cosmology. He fell in love with the idea of high-order interactions, of emergent properties and ended up earning a PhD on complex networks at Imperial College London in 2012. Theoretical approaches never stopped fascinating him, and he continues this research today working at the interface between complex systems and algebraic topology. His research spans the analysis of neuroimaging data and AI systems with topological techniques, the formalization of cognitive control models with tools of statistical mechanics and network theory, and the study of the predictability of socio-technical systems.
Andrea Luppi, McGill University, Canada
Transitions between cognitive topographies: contributions of network structure, neuromodulation, and disease
To explain how the brain orchestrates information-processing for cognition, we must understand information itself. Importantly, information is not a monolithic entity. Information decomposition techniques provide a way to split information into its constituent elements: unique, redundant, and synergistic information. In this talk I will outline how disentangling synergistic and redundant interactions is redefining our understanding of integrative brain function and its neural organisation. To explain how the brain navigates the trade-offs between redundancy and synergy, I combine multimodal and multi-species evidence integrating the structural, molecular, and functional underpinnings of synergy and redundancy; their roles in cognition and computation; and how they might arise over evolution and development. Overall, disentangling synergistic and redundant information provides a guiding principle to understand the informational architecture of brain and cognition.
Andrea Luppi studied philosophy, neuroscience, and cognitive science at the University of Oxford and Harvard University, before obtaining a PhD in neuroscience from the University of Cambridge and the Alan Turing Institute. He is currently a Molson Neuro-Engineering Fellow and Neuro-AI Postdoctoral Scholar at the Montreal Neurological Institute of McGill University in Montreal, working in the group of Prof. Bratislav Misic. Andrea’s work investigates how cognition and consciousness arise from the complex interplay of network structure and brain dynamics across scales. To this end, he combines approaches from information theory, network science and whole-brain computational modelling to investigate the dynamics and connectivity of the brain across pharmacological and pathological perturbations. His long-term goal is to understand what makes artificial and biological neural networks capable of consciousness and intelligence.
Fernando Rosas, University of Sussex, UK
Formal approaches to emergence: theory, practice, and opportunities for brain analyses
Emergence is a profound subject that straddles many scientific scenarios and disciplines, including how galaxies are formed, how flocks and crowds behave, and how mental activity arises from the orchestrated activity of neurons. At the same time, emergence is a highly controversial topic, surrounded by long-standing debates and disagreements on how to best understand its nature and its role within science. This talk presents formal approaches to characterise emergence as way to advance these discussions. Specifically, we argue that these approaches make emergence useful for empirical practice, giving researchers rigorous frameworks to guide discussions and advance theories, and also quantitative tools to establish conjectures about emergence and then test them on empirical data. We provide an overview on the theory and practice of formal approaches to emergence, and highlights the opportunities they open for studying the relationship between the brain and the mind. The talk presents illustrative examples of the application of principles of emergence to practical data analysis, discussing several key interpretation issues and avenues for future work.
Fernando Rosas works as lecturer at the University of Sussex and research fellow at Imperial College London and the University of Oxford. His work is focused on the development of conceptual and computational tools to better understand the collective dynamics of complex systems. His background includes studies in mathematics, physics, philosophy, and music composition.
Huifang Wang, Aix-Marseille Université, France
Virtual brain twins in epilepsy and beyond
In this talk, we will have the definition of virtual brain twins, its key element and relationship between these elements. Then we will see the concept of personalized whole-brain network models. I will give a solid example in terms of the personalized whole-brain network modelling in epilepsy, which we call virtual epilepsy patients (VEP). I will explain how to use VEP to estimate epileptogenic networks, plan virtual surgeries and stimulations. In the end, we can extent this personalized whole-brain network modelling in other uses, such as healthy ageing, Alzheimer’s, multiple sclerosis, Parkinson’s and psychiatric disorders.
Dr. Huifang Wang is a neuroscientist who works with Dr. Viktor Jirsa in INS, an AMU and INSERM institute. Her current research is interested in personalized whole brain modelling (digital twins) in neurology. Led by Viktor, they have built a virtual epileptic patient pipeline for the diagnosis and treatment of epilepsy. Now we are on the journal to digital twins for the dignosis, treatement and prognosis for epilepsy and other brain disorders.
Henrik Jensen, Imperial College London, UK
Brain dynamics: Shadows of long-range correlations and criticality.
In statistical mechanics a system is said to be in a critical state if the correlation length is infinite. This implies the lack of specific characteristic scales, and the resulting self-similar structure of the system is characterised by power laws. The structure of brain dynamics shows many aspects of this type of critical behaviour. This has led some authors to look for ways to quantify “how critical” a particular brain is in a particular situation. We will mention why the breaking of scaling relations is an unlikely biometric measure of the degree of the brain’s criticality. We will suggest a better alternative is to focus on monitoring correlations in space and time combined with information theoretic measures.
Henrik Jensen is an expert on the statistical mechanics of complex systems. His most recent book is "Complexity Science: The Study of Emergence" (Cambridge Press) which presents complexity science as the organised scientific study of collective emergent behaviour. He has worked on the dynamical properties of condensed matter systems and developed the Tangled Nature model of evolving ecosystems. An approach of broad relevance to evolving adaptive complex systems. His work together with Paolo Sibani on record dynamics and its relevance to a surprisingly broad range of complex system, including macroevolution and the dynamics of ant colonies, have attracted interest from researchers from biology to materials science. Recently he has focused on brain dynamics and structure by analysing fMRI and EEG data making use of approaches inspired by statistical mechanics and information theory.
Fran Hancock, King's College London, UK
Metastability demystified – praeteritum, present, futurum
Healthy brain functioning depends on balancing stable integration between brain areas for effective coordinated functioning, with bursts of desynchronisation to allow subsystems to reconfigure and express functional specialisation. Metastability, a concept originated in statistical physics and dynamical systems theory, has been proposed as a key signature that characterises this balance. Building on this principle, the neuroscience literature has employed markers of metastability to investigate various aspects of brain function including cognitive performance, healthy ageing, meditation, sleep, responses to pharmacological challenges, and to characterise psychiatric conditions or disorders of consciousness. However, this body of work often uses the notion of metastability heuristically, and sometimes inaccurately, making it hard for the uninitiated to navigate the vast literature, interpret findings, and foster further development of theoretical and experimental methodologies. In this presentation I will provide an abridged review of metastability and its applications in neuroscience, recall its scientific and historical foundations, explore some practical estimators used in empirical data, and suggest ways forward for future developments.
Fran is a visiting research fellow at the IoPPN in King’s College London collaborating with Prof. Federico Turkheimer at the Centre for Neuroimaging Studies. Following a BSc in applied mathematics and physics, and an MBA, Fran worked for many years as a management consultant in the pharmaceutical industry before taking a senior executive role in the R&D organization of a global consumer goods company. She returned to academic pursuits completing a BSc in psychology, an MSc in Applied Neuroscience, and most recently, a PhD in applied neuroimaging at King’s. Given her diverse experience and academic interests, Fran strives for conceptual integration of concepts and theories that sometimes remain isolated in specific domains such as cognitive psychology, dynamical systems theory, information theory, or stochastic processes. One such concept is metastability for which Fran continues to investigate empirically, theoretically, and computationally.
Giuseppe de Alteriis, King's College London, UK
A Framework for brain-wide network dynamics: from calcium imaging to fMRI
Dynamic Functional Connectivity is a growing field of research in neuroimaging. In my talk, I will propose a new framework to calculate, analyze and interpret dynamic connectivity patterns based on their mathematical structure. I will explain why this framework has the level of abstraction necessary to analyze brain-wide network dynamics at different spatio-temporal scales, showing evidence from widefield calcium imaging and fMRI in rodents. Finally, I will show in different neuroimaging datasets how some of the brain-wide network properties are disrupted with ageing and in psychiatric disorders and how this is related to cognitive abilities.
Giuseppe is a LIDo (London Interdisciplinary Doctoral Programme) PhD student at King's College London and UCL. His research is focused on translating ideas from complex systems theory and network theory to the brain. His ambition is to work at different brain scales, from 2p imaging of single neurons in rodents to human fMRI, in order to investigate the neural basis of brain wide network dynamics.
Pedro Mediano, Imperial College London, UK
Entropy and Complexity in neural activity
There are many ways of measuring complexity in experimental data, each of them with different strengths and weaknesses. Here we will focus on the information-theoretic construct of entropy, which quantifies complexity as the degree of disorder in a system. We will cover several empirical results linking entropy in the brain with level of consciousness, and present new results exploring this relationship. Finally, we will finish by discussing new open-source tools available to estimate entropy from neural time series and show that these yield rich, meaningful analyses of experimental data.
Pedro holds a degree in Physics from the University of Valencia, a PhD at Imperial College and a postdoc at the University of Cambridge, where he worked on consciousness and complexity. Now, he is a lecturer at the Department of Computing, Imperial College London.
Broadly speaking, his research lies at the intersection between consciousness science, information theory, complexity science, and machine learning. He strives to ground his research in solid mathematical and theoretical grounds, and whenever possible he publishes open-source software to make his research accessible and reproducible.