The Systems Neuroscience program follows the approach of systems theory in understanding the brain. The aim is to provide students a view of the brain as a whole via unfolding, at least in part, its immense complexity. This is a major challenge of all time, but the right answer should be one that can integrate actual knowledge. As we are in the fortunate period of time when high performance tools (both hardware and software) and large datasets are getting more and more available, systems thinking is inevitable in brain research. Therefore, throughout the course students will learn how different approaches - reductionist, holist and functionalist - are all useful and necessary in understanding the brain.
In one way the course is structured by introducing the students the different levels of organization all being complex systems themselves. We start with molecular machineries at the subcellular level (course A.I.) then turn into the cellular level by learning why the neuron is considered as the unit of brain organization (course A.II.). In the next step it is shown how the milliards of neurons make up the cerebral cortex and how this evolutionarily new structure can perform diverse cognitive and other functions (course A.II.). Finally, whole brain functions and functioning will be approached via its role in behavior (course C.I.).
In other way, it will be shown how the brain functions at lower organizational levels such that synaptic and neuronal populations of different structures (course B.I.) as well as a whole (courses B.II., III., and C.I.). Such holistic approaches have high importance in recent efforts made in deciphering the neurobiological basis of neuropsychiatric and neurological diseases (courses B.II. and III.). Finally, with the closing series of lectures (course C.II.) we aim to provide tools, rules and examples, which help integrating knowledge acquired throughout the courses and also provide an outlook whereby the brain can be compared to other complex systems. Another notable feature of the Systems Neuroscience program is its interdisciplinary nature: it will introduce the students into several state of the art methods both experimental (molecular biology, cellular and extracellular physiology, different kinds of imaging) and theoretical (data and network analyses, simulation and modelling).
The course will start with an introductory about the history and culture of the geopolitical region thought as Central Europe.
Central European Culture
By Eszter Greskovics
Archivist, art historian at Artpool Art Research Center, Budapest
The course starts with a general overview of Central Europe, by explaining briefly the history and the changing meaning and perception of the term “Central Europe”; the discussion includes geographical introduction to the region as well, i.e. which countries are considered to be Central European (and why).
Following the introductory part Medieval and Early Modern history of the region is discussed. This part includes the role of religion in the region with additional examples of the art and architecture of the period.
17th century to 19th century, Austro-Hungarian Empire, WWI
The first half of the lecture is dedicated to the road that led to the industrial revolution and to the national revolutions of the 19th century throughout Europe and the revolutions themselves. The main focus remains on Central Europe: nationalism and the spring of nations in Central Europe. This part also includes examples in the form of art, architecture and literature.
The history and importance of the Austro-Hungarian Empire is the focus of the second half of the lecture. This includes architectural, literary and gastronomic examples (e.g. similarities that are still present). WW1 is also one of the main topics of this lecture together with the disintegration of the Empire and the new borders following the Treaty of Trianon.
WW2, Central European film
The interwar years are discussed briefly at the beginning of the lecture but the main focus is WW2, including the Holocaust and its significance in the Central European countries. The lecture is supplemented with visual aids: photos, video excerpts, etc.
To counterbalance the serious topic of the lecture a short introduction is included to Central European film (e.g. Czech New Wave). This also connects the lecture to the next one.
The last day of the course is dedicated almost solely to the period after WWII. First of all, the effects of the war are discussed and the case of the “two Central Europes” (i.e. East and West Germany) is introduced. Second of all, Communism and the notion of the Iron Curtain (and the life behind it) are presented. Finally, at the end of the course the system changes of 1989 in the region are highlighted. This section includes video illustrations about the fall of the Berlin Wall and/or about everyday life in the Soviet Bloc.
Guided tour in the Dohány Street Synagogue (the largest synagogue in Europe) and/or in the Jewish quarter of Budapest.
Guided tour in Hospital in the Rocks (Sziklakórház, http://www.sziklakorhaz.eu/en/) Sightseeing boat tour (e.g. http://www.legenda.hu/en)
By Gábor Juhász, PhD, DSc
Laboratory of Proteomics, Institute of Biology, Eötvös Loránd University, Budapest
Psychiatric and neurodegenerative diseases are the most challenging questions in brain research. Since no causal therapy for brain diseases exists and a large portion of the society is involved in brain diseases or treating of diseased relatives. Actually big governmental projects support brain sciences because of increasing social pressure. One of the most important parts of the rapidly changing knowledge about how the brain works is the changing view to the synapse, the information transmitting element of neurons. Classical model of synapse describes synapse as a signal conducting element between neurons converting action potential to a chemical signal back and forth. After a short review of classical synapse model we turn to the systems biology fuelled model of synapse. Systems biology approaches the synapse as a complex molecular machine and concerns with protein interaction network based explanations of synaptic functions. Summarizing the molecular composition of pre- and post- synaptic membrane at protein interaction level, we give scope to how synaptic molecular machine control synaptic transmission via positioning neurotransmitter release sites against post-synaptic receptors and in turn induces long term potentiation. Also we discuss how synaptic molecular machinery could be printable by the transmitted information and what the role of synapse in memory trace formation is. Finally we give the basic ideas and facts about recovery and stability of synaptic protein machine and the role of synaptic energy production made by synaptic mitochondria will be discussed. Our aim is to demonstrate how systems biology can give a novel insight into synaptic function and enhance our knowledge about information transfer and processing in the brain. We also demonstrate how pharmaceutical industry can find novel synaptic protein targets using unbiased research strategy applied on the synaptic proteome of healthy and diseased synapses. It allows to have a scope to the molecular pathomechanisms of most common brain diseases. The lecture intends to convince students about the power of using protein network complexity based molecular machine principle in understanding synaptic functions. Several data shown in the lecture are collected in our own laboratory.
Genomic analysis of single neurons
By Árpád Dobolyi, PhD, DSc Department of Anatomy, Histology and Embryology, Semmelweis University, Budapest Laboratory of Proteomics, Institute of Biology, Eötvös Loránd University, Budapest
Recently, the single nucleic acid analysis method development resulted in a change in our understanding of phenotype at the cellular level. Single cell transcriptome analysis is based on cell harvesting followed by linear amplification of cDNA converted form mRNA samples. The DNA analysis of the amplified samples can be done by PCR, digital PCR and next generation sequencing. The result is not only the sequences of gene products but also the copy number of a different mRNAs in a cell. The technology is able to follow the copy number down to 2-3 copies. Copy number data revealed that cells are individual in great extent so the traditionally established neuronal phenotypes are only simplified clustering of the cells. The phenotype is dynamic and changing in a short time scale so a real cell has an equi-phenotype space, in which the cell is continuously moving because of dynamic fluctuations in protein transcription regulated by actual needs. There are transitional phenotype spaces, through which one cell can be converted to another one, like stem cells and induced pluripotent cells iPCs. When the phenotype range of a cell gets very limited, the cell is to die. Such a new scope to the genomic control of cell functions opened a new era in neurobiology because of better understanding of the protein-cell function relations. The lecture is based on studies made at UPENN and also includes some of our own data for demonstration of this exciting novel field of neuroscience.
Single neurons and beyond
By Balázs Ujfalussy, PhD
Lendület Laboratory of Neuronal Signaling, Institute of Experimental Medicine (IEM) of the Hungarian Academy of Sciences, Budapest, Hungary
Our cognitive abilities - making sense of the images projected to our retina, storing those images in our memory for years and recalling them when they are required to make an appropriate decision - all rely on the concerted interaction of millions of neurons composing the brain. The aim of this course is to understand how their biophysical properties render single neurons highly efficient information processing units so that when connected together, they create the most powerful computational device known to date.
We will start by a concise introduction to the biophysical mechanisms responsible for generating neuronal activity which provides a basis for understanding signal integration and propagation in single neurons. Next, we will introduce various approaches to model signal processing in neurons, ranging from highly detailed descriptions to greatly simplified caricatures, each useful for studying different aspects of neuronal function. We will compare these models based on two criteria having special importance from a systems perspective: their ability to predict the response of the neurons to natural - sensory or synaptic - stimuli; and their potential to identify principles of neuronal information processing. Finally, we will use specific examples to illustrate how particular features of individual neurons contribute to the computation (sensory processing, memory or decision making) performed by the system.
Neocortex: from structure to function
by László Négyessy, PhD
Complex Systems and Computational Neuroscience Group, Wigner RCP, Hung. Acad. of Sciences, Budapest and Department of Anatomy, Histology and Embryology, Semmelweis University, Budapest
The cerebral cortex is evolutionary the youngest and the most complex structure of the brain. At a first sight it seems as a thin, homogenous, continuous and crinkled sheet wrapping the rest of the brain. Notably, this sheet is responsible for the highest level neural integration including cognitive functions. However, the cerebral cortex also exhibits remarkable functional specificity. Understanding the neurobiological mechanisms of these two features, integration and specificity, which implies segregation, ultimately helps understanding how the behaviorally relevant structures and pathways are activated in the cerebral cortex. In this course we will overview the current knowledge about the neurobiological basis of cortical integration and segregation via complex systems approach.
Multiscale organization will be reviewed by decomposing the cortex into building blocks from neurons through columns or modules to cortical areas, and showing how the components interact and form higher level organizations. Through the course of such nested networks from microcircuits up to the large scale cognitive networks emerging features of cortical structure (especially those regarding connectivity), function (i.e. dynamics) and their relationship will be discussed. We end up at the large scale network of cortical areas and identify specific cognitive neural architectures as sub-networks formed via integrating distributed processing.
By Zoltán Somogyvári, PhD
Complex Systems and Computational Neuroscience Group, Wigner RCP, Hungarian Academy of Sciences, Budapest
The neural system generates electric field, measurable on different scales, among which the most widely known is called electroencephalogram (EEG) and can be measured by macroscopic electrodes attached to the scalp.
In this course, first the generation process underlying the neuro-electric phenomena will be described, starting from the microscopic scale by describing the membrane potential of a single neuron, up to the macroscopic scale of the EEG. Then, up-to-date mathematical analysis methods will be introduced, which makes possible the inference of fine details of neural structures and their activity, based on the recorded electric signals.
We will see, how to determine the anatomical structure from which the measured electric activity is originated, based on solely the electric signals.
The study on the source localization methods and their application to the microscopic field potential of single neurons will reveal fine details of inputs and outputs of the neurons, which can not be measured by any other technique today.
The microscopic electric and optical measurements reveal interesting co-occurrences or correlations between specific temporal and spatial structures of activations, waves and oscillations of electric and metabolic activity. However, in order to understand the dynamics of the investigated neural system, we should demonstrate not only the correlations but the causal relationship between the observed patterns. Showing the causality is a much more difficult task, than demonstrating the correlations. Thus, new mathematical tools and their applications for detecting causal relationships and information flow will be presented on the course.
Neural rhythms: normal and pathological
By Dániel Fabó MD PhD
Department of Functional Neurosurgery and Department of Epilepsy, National Institute of Clinical Neuroscience, Budapest
The brain is a very complex oscillator. This fact has been obvious from the very first electroencephalographic recording of Berger’s, when the alpha oscillation was recorded. Since than the functional implications of brain oscillations changed significantly, from being idling rhythms to harboring essential functions. Parallel with the physiological studies, investigations of the pathological brain states resulted numerous other oscillations, the registration of which became basic tools of classical neurological diagnostics. The idea of the underlying mechanisms, how pathological brain oscillations are organized, evolved also significantly. Now many think that derailed physiological oscillators produce the altered rhythms, chaining together the investigation of pathologies with the understanding of physiological functions. It is studied intensively if these failed rhythms are causes or consequences of the diseases and the symptoms. Regardless which situation is the case, a new therapeutic strategy has been emerged from these thoughts called neuromodulation. Through these techniques new rhythms are delivered into some parts of the nervous system of various patients resulting the easing or elimination of the symptoms.
During this course we set sail adrift these waves to get better insight into the normal and abnormal functions of the brain. We proceed through wake and sleep oscillations to the pathophysiological features of epilepsy, movement disorders or chronic pain. We will see how these oscillations can be recorded in humans at various scales, beside the bed or within the operating room and we will glance at the exciting horizons opened by neuromodulatory techniques such as deep brain stimulation, transcranial electric and magnetic stimulation, or peripheral nerve stimulation.
Brain imaging: from normal to pathological
By Lajos Kozák, MD, PhD
MR Research Center, Semmelweis University, Budapest
Blood-oxygenation level dependent (BOLD) functional MRI (fMRI) provides a highly flexible, high spatial resolution, and non-invasive and safe means for measuring and describing neural activity in humans. Due to its favorable properties it became a widely used method for investigating and describing functional networks in healthy individuals and patients.
Functional MRI started off as a brain mapping tool, i.e. relatively simple paradigms were used to localize brain areas involved in various cognitive functions. Later on, as the method evolved and became an important tool of cognitive science, paradigms became increasingly complex and the description of functional networks and the investigation of network hierarchies became available. Recently, data-driven approaches lead to the era of paradigmless or resting-state fMRI which in conjunction with diffusion tensor imaging (DTI, an MRI-based tool for the in-vivo non-invasive mapping of white matter structure) became the most important tool for large scale whole brain analysis of functional and structural connectivity networks of the brain.
The course starts with an introduction of the physical and physiological background of fMRI and DTI, then we will continue with the basics of data acquisition and classical generalized linear model (GLM) based analysis, paradigm design for brain mapping, dynamic connectivity (dynamic causal modelling, DCM) analysis and data-driven methods (independent component analysis, ICA), with a focus on basic research and clinical applications. We will demonstrate and discuss relevant publications related to fMRI- detectable differences between healthy and diseased brains, and the cognitive and possible structural correlates of such differences.
Special focus will be given to functional brain mapping as a tool for pre-surgical evaluation and decision-making in brain tumor and epilepsy patients. Lectures will be accompanied by hands-on practicals of data acquisition, data pre-processing and fMRI analysis using free or open access research tools.
Statistics and the Brain
By Gergő Orbán, PhD
Computational Systems Neuroscience Lab, Wigner RCP, Hungarian Academy of Sciences, Budapest
Perception relies on internal models to interpret stimuli coming from the external world. These internal models represent our knowledge about the dependencies and interplay of the features of the environment. Characterisation of these models is challenging and requires advanced techniques. a particularly promising way to understand the internal models applied by the nervous system is to identify the computational problem these internal models need to solve. The nervous system is bound to use available information efficiently in order to make decisions or execute motor plans. In a general setting the information collected by our senses is incomplete, noisy, or ambiguous and computational strategies applied by the nervous system need to overcome these limitations as well as possible. We adopt a Bayesian perspective to construct ideal observer models which define the optimal strategy to interpret stimuli. Whether the generation of stimuli involved a stochastic component or not, a Bayesian treatment assumes a probabilistic process underlying the stimulus. We will explore how humans adopt these optimal strategies, how computations can be identified by psychophysical experiments, how internal models characteristic to each individual can be revealed and what approximations are made once computations become intractable.
Complexity of the brain
By Péter Érdi, Henry R Luce Professor
Center for Complex Systems Studies, Kalamazoo College, Kalamazoo, Michigan , USA Complex Systems and Computational Neuroscience Group, Wigner RCP, Hungarian Academy of Sciences, Budapest
It is often said in colloquial sense that the brain is a prototype of complex systems. A few different notions of complexity may have more formally related to neural systems. First, structural complexity appears in the arborization of the nerve terminals at the single neuron level, and in the complexity of the graph structure at network level. Second, functional complexity is associated to the set of tasks being performed by the neural system. Third, dynamic complexity can be identified with the different attractors of dynamic processes, such as point attractors, closed curves related to periodic orbits, and strange attractors expressing the presence of chaotic behavior.
The course gives a bird's eye perspective on (i) Dynamical system's theory to the brain (ii) Computational neurology and psychiatry (iii) Computational neuropharmacology (iv) Network-theoretical approach to the brain