Terrence J. Sejnowski - US grants

The PI seeks support for an investigation of a novel technique to quantify the changes in intracellular concentrations of calcium that are associated with the activity of hippocampal neurons involved in learning and memory. Specifically, calcium- sensitive fluorescent dyes will be mircroinjected into selected neurons and monitored using a dual-wavelength spectrofluorimeter and scanning confocal microscope following manipulation of neuronal activity, such as high-frequency tetanus to produce long term potentiation. The proposed research is a gamble and there is a minimum of preliminary data to allow an easy judgement. However, the investigators have made excellent progress in analyzing calcium changes in cultured cells. The difficulty is in extending these procedures to the tissue slice. The potential for a significant outcome seems sufficient to justify the risk. The ability to make high-resolution measurements of calcium changes in hippocampal neurons will have an impact on many areas of neuroscience, including computational psychology, neural net modeling, and electrophysiology. This work has the potential to move neuroscience significantly closer to an understanding of the synaptic molecular mechanisms that underlie learning and memory.

The long-term goal of this research is to understand the overall physiological, anatomical, and functional organization of the mammalian hippocampus and hippocampal formation. In this proposal the focus is on modeling the electrical and chemical signals that occur in single CA1 pyramidal neurons based on experimental data. These models are being developed in parallel with ongoing experimental research in the same laboratory. The purpose in developing these models is tho supplement and extend our understanding of biophysically complex neurons as computational elements and to aid our intuition in the design and interpretation of experiments. The specific aims of the project are: (1) to characterize the passive electrical properties of CA1 pyramidal neurons using whole-cell clamp and convention intracellular impalement recordings, in conjunction with cellular three-dimensional reconstruction and computer modeling; (2) to develop a more complete and realistic computer simulation of processes occurring in dendritic spines at the biophysical level, including the effects of intracellular diffusion and buffering of ions; (3) to develop more realistic models of the excitability of CA1 pyramidal cells, based on existing voltage clamp, histochemical, morphological and current- source density data; and (4) to use these simulations to determine the critical parameters for associativity and specificity for long-term potentiation (LTP) and long-term depression (LTD) in hippocampal neurons. The proposed research will use modeling tools that have been recently developed for simulating detailed models of dendritic trees with hundreds of compartments. The CABLE simulator developed by Hines and Moore will be modified to include the diffusion, binding, uptake and release of calcium and other diffusible second messengers. Simulations of calcium entry into dendritic trees through voltage-sensitive calcium channels and NMDA receptors will be compared directly with measurements made using calcium-sensitive dyes and confocal microscopy in the same laboratory. An accurate model of electrical and chemical processing in hippocampal neurons would be of benefit for many other investigations of hippocampal function and dysfunction, including epileptogenesis and memory impairments brought about by selective anatomical lesions and physiological imbalances.

9408680 Sejnowksi Recently a new field of engineering has emerged, referred to as neuromorphic engineering, that is based on the design and fabrication of artificial neural systems, such as vision chips, head-eye systems, and roving robots, whose architecture and design principles are based on those of biological nervous systems. A two-week workshop on neuromorphic engineering will be held to bring together young investigators and more established researchers from academia with their counterparts in industry and national laboratories, working on both neurobiological as well as engineering aspects of sensory systems and sensory-motor integration. Formal lectures will be given, but the primary focus will be hands-on experience with research tools for all participants. The workshop will serve as a bridge between the engineering world of artificial neural systems and the neuroscience community. The workshop will provide an environment for intensive interactions between members of these two communities - merging engineering principles with experimental results from neuroscience. The workshop will also serve as a prototype for the development of a similar yearly summer workshop. The interaction of these two disciplines should have significant impact both on the development of new technologies (new artificial neural systems) and on our understanding of how the nervous system is designed.***

9511637 Sejnowski Recently a new field of engineering has emerged, referred to as neuromorphic engineering, that is based on the design and fabrication of artificial neural systems, such as vision chips, head-eye systems, and roving robots, whose architecture and design principles are based on those of biological nervous systems. A two-week workshop on neuromorphic engineering will be held to bring together young investigators and more established researchers from academia with their counterparts in industry and national laboratories, working on both neurobiological as well as engineering aspects of sensory systems and sensory-motor integration. Formal lectures will be given, but the primary focus will be hands-on experience with research tools for all participants. The workshop will serve as a bridge between the engineering world of artificial neural systems and the neuroscience community. The workshop will provide an environment for intensive interactions between members of these two communities - merging engineering principles with experimental results from neuroscience. The interaction of these two disciplines should have significant impact both on the development of new technologies (new artificial neural systems) and on our understanding of how the nervous system is designed. ??

Humans and other animals use visual 'looming' of a stimulus to detect change in distance of a stimulus in the depth of the visual field. It is unclear how such visual cues drive neural signals that guide appropriate behavioral responses such as approach or avoidance for such a stimulus. This project uses an insect, the moth Manduca, which hovers in front of flowers while feeding in flight, as a simpler system for experimentation. A combined approach links behavioral experiments to physiology and to modeling, to experimentally test alternative hypotheses about which visual cues are relevant for guiding behavior, as a basis for developing computational models with realistic biological parameters.
Results will be important beyond insect vision, for understanding depth detection and obstacle avoidance by visual mechanisms in general, and for developing useful machine vision and guidance systems in robotics. This project also provides excellent cross-disciplinary training of a postdoctoral woman neuroethologist in an exceptionally strong environment for computational neuroscience.

The primary goal of this training program in cognitive neuroscience is to provide advanced graduate students and postdoctoral fellows with interdisciplinary research training in Cognitive Science, Neuroscience and Computation in preparation for a career in cognitive neuroscience. The training program will emphasize interdisciplinary training that involves collaborative research between laboratories using several different techniques. Fifteen faculty at The University of California at San Diego (UCSD) and The Salk Institute for Biological studies will participate in the training program, which builds on eight years of experience with a highly successful Center for Cognitive Neuroscience. Graduate students will be drawn from a large pool of high-quality applicants who are enrolled in the Departmental of Cognitive Science, the Neurosciences Graduate Program, and a new Computational Neurobiology Program in the Department of Biology at UCSD, as well as from other cognate departments including Philosophy, Psychology, and Psychiatry. The resources of several Organized Research Units at UCSD are also available for graduate and postgraduate research training including the Center for Research in Language, the Institute for Neural Computation and the Center for Human Information Processing. The advisory committee that will supervise the training program consists of T. Albright, E. Bates, U. Bellugi, S. Hillyard, M. Kutas, T. Sejnowski, L. Squire, and D. Swinney. The advisory committee will meet with the chairs of all the relevant graduate programs at UCSD to develop a coordinate infrastructure for training in cognitive neuroscience. In addition to the predoctoral training, support for postdoctoral research will focus on new research projects that develop new approaches toward understanding higher brain function. All of the major research areas are represented by the participating faculty on this training grant, including vision, memory, attention, language, sleep development, and neurophilosophy. Training will be provided in a wide variety of methods including electrophysiology, psycholinguistics, functional magnetic resonance imaging computational modeling, and developmental neurobiology. The training in basic research will directly involve studies of humans with mental health problems, including aphasia, autism, Williams syndrome, Downs syndrome and sleep disorders.

General Monte Carlo Simulation of Subcellular Biochemical Signaling, Phase II

At the molecular level, cells are highly organized and disorganized at the same time. Many of the protein building blocks of cells form highly organized structures used for a variety of purposes, such as communication between cells and holding cells together. On the other hand, when small signaling molecules are released they can bounce around in random directions in a seemingly disorganized way, until they find a binding site on another molecule like a protein or section of DNA. These two forms of organization and disorganization are at the heart of a new computer program called MCell. This program follows individual molecules as they bounce around a realistic three-dimensional world populated with large proteins and other highly organized structures that accurately mimic the shape and contents of a cell. The purpose of this grant is to expand MCell's ability to simulate the molecular processes that take place inside cells and the communication that occurs between cells.

The research will be carried out jointly between The Salk Institute (T.J. Sejnowski and T.M. Bartol), Carnegie Mellon University (J.R. Stiles), and Cornell University (E.E. Salpeter). Over 21 laboratories around the world are currently using MCell to study synaptic transmission at the nerve-muscle synapse and a wide variety of synapses in the brain, as well as biochemical processes in cells of other organs such as the liver. The grant will also provide for the development of a graphical user interface, which will make it easier and faster to use the program, and extensive documentation to make MCell more accessible to new users. One of the most important new features of MCell is its ability to model the complex three-dimensional structure of real cells, which can be highly convoluted and difficult to visualize. A new graphical tool for reconstructing, editing, and visualizing three-dimensional cellular structures at the level observed only with electron microscopes will be incorporated into the next generation of MCell. This will also make it possible for MCell users to deposit their reconstructed models into a web-based archive, which will promote the accumulation and free exchange of models among researchers. There will be several hundred additional MCell users by 2003. MCell simulations may be run on a wide range of platforms, from hand-held calculators, to standard personal computers, to large-scale supercomputers. The additions to MCell that will be supported on this grant will greatly expand the range of problems that can be examined using MCell, bringing closer the day when it will be possible to simulate the functions of an entire cell in a computer.

With the addition of the graphical user interface, a major aim of a future phase of development will be to make MCell available as a teaching tool both for students who are beginning their science education, to help them visualize the properties of cells, and for advanced students, who can use the program as a means to explore the complex and sometimes counterintuitive world at the molecular level. For more information about MCell and computer-generated images of what cells might look like at the molecular level, see http://www.mcell.cnl.salk.edu or http://www.mcell.psc

9987614
Terry Sejnowski - University of California, San Diego
IGERT: Graduate Training Program in Computational Neurobiology

This Integrative Graduate Education and Research Training (IGERT) award supports the establishment of a multidisciplinary graduate training program of education and research in computational neurobiology. The goal is to train a new generation of scientists and engineers with a broad range of scientific and technical skills who are equally at home measuring large-scale brain activity, analyzing the data with advanced computational techniques, and developing new models for brain development and function. This integrative training program is centered in the Department of Biology at UCSD and the Salk Institute, but includes faculty members from physics, chemistry, psychology, cognitive science, electrical engineering, computer science, and mathematics, as well as from biology and neuroscience. The training program will give all students hands-on experience in a wide range of advanced experimental and computational techniques through collaborative research between laboratories, industrial internships, and the opportunity to pursue research abroad. The faculty will participate in outreach programs to encourage and prepare underrepresented minorities for a career in computational neurobiology. Research areas in the training program include: (1) synaptic growth and plasticity; (2) neural dynamics; (3) neural population coding; (4) visual perception and memory; (5) stochastic learning algorithms; and (6) functional brain imaging.

IGERT is an NSF-wide program intended to meet the challenges of educating Ph.D. scientists and engineers with the multidisciplinary backgrounds and the technical, professional, and personal skills needed for the career demands of the future. The program is intended to catalyze a cultural change in graduate education by establishing new, innovative models for graduate education and training in a fertile environment for collaborative research that transcends traditional disciplinary boundaries. In the third year of the program, awards are being made to nineteen institutions for programs that collectively span all areas of science and engineering supported by NSF. The intellectual foci of this specific award reside in the Directorates for Biological Sciences; Computer and Information Science and Engineering; Social, Behavioral, and Economic Sciences; Mathematical and Physical Sciences; Engineering; and Education and Human Resources.

DESCRIPTION (Verbatim from the Applicant's Abstract): It is now possible to continuously record the brain's electromagnetic field from sixty or more scalp locations and concurrently to record brain blood flow or oxygenation level changes at tens of thousands of mm3 sized brain volume elements. Methods are needed to analyze this wealth of data and to separate out machine noise and physiological artifacts to examine functionally independent brain processing systems. The aims of the proposal are threefold: (1) To test the feasibility of concurrent functional magnetic resonance imaging (fMRI) and high density encephalographic (EEG) recording during performance of simple and complex cognitive tasks; (2) To test applications of Independent Component Analysis (ICA) to concurrent fMRI and EEG data, first separately, by comparing the resulting fMRI source distributions with EEG source distributions derived from the ICA results using currently available EEG source localization approaches, and then jointly, by directly comparing the time courses of changes in fMRI activation and in the EEG frequency power spectrum; and (3) To develop, test, document and distribute a software toolbox, based on the widely-used MATLAB signal processing environment, for carrying out the analyses we envision and visualizing the results. The tools, which will be written for the widely available MATLAB environment, will allow researchers to decompose concurrently or separately recorded EEG and fMRI data into spatially and/or temporally independent components, and to evaluate their relationships to concurrent measures of task performance or other physiology and behavior. The analysis of electrophysiological and functional imaging data in both time and time/frequency domains will allow researchers to determine and evaluate relationships between ongoing or event-related changes in electrophysiology, hemodynamics and behavior and perform exploratory and hypothesis drive analyses on human brain data from both basic and clinical studies of human brain and cognitive function.

DESCRIPTION: (Adapted from the Investigator's Abstract): The intrinsic cellular, synaptic and network mechanisms underlying cortically-initiated electrographical seizures will be explored with electrophysiological experiments in vivo and in computational modeling I studies. The electrographical seizures observed in cat neocortex resemble partial and generalized epileptic | seizures consisting of spike-wave (SW) or polyspike-wave (PSW) complexes at 1.5 about Hz and fast runs at 10 15 Hz. We will study (a) specific cellular and network mechanisms responsible for the onset, maintenance and I termination of the paroxysmal seizures; (b) the specific and common properties of different types of | electrographical seizures; (c) the conditions responsible for transforming spatially localized paroxysmal activity into generalized seizure. Compartmental models of excitatory and inhibitory neocortical neurons and their interactions will be used to examine specific hypotheses for the mechanisms underlying multiple intracortical processes preceding, accompanying and following electrographic seizures and to generate predictions that can be tested experimentally. These models may also be useful for preliminary screening of antiepileptic drags. The methods that will be used in these studies include (a) in vivo multisided intracellular and l field potential recordings from anesthetized and behaving animals to probe specific intracellular and synaptic changes during seizure; (b) recently developed independent component analysis (ICA) applied to the I electrophysiological data to study spatio-temporal properties of paroxysmal activity; (c) computational modeling based on detailed description of the intrinsic properties of individual neurons and their synaptic interconnections. The modeling techniques are aimed at understanding paroxysmal seizure mechanisms by simultaneous and independent observations of all contributing factors, some of which are difficult to study experimentally. Electrophysiological recordings and preliminary analysis of the experimental data will beconducted in Laval University (Canada) and detailed computational analysis based on ICA methods and modeling techniques will be conducted in the Salk Institute (USA). These studies may lead to new treatments for epilepsy based on comparisons between models for the normal and diseased states of the cortex.

Automatic Analysis of Spontaneous Facial Expressions
Abstract
The goal of this project is to develop computer systems for automatic analysis of spontaneous facial expressions, with a focus on the scientific study of the role of facial expressions in deception. A state-of-the-art digital video database of spontaneous facial expressions will be developed. This database will be hand-coded by behavioral scientist experts on facial expressions. This database will be used to develop an array of software tools for automatic analysis of facial expressions from video sequences. These tools will be developed by machine perception scientists in close collaboration with behavioral scientists and will be evaluated and refined for application to the scientific study of facial expressions.

The machine perception community is in critical need for standard video databases to train and evaluate systems for automatic recognition of facial expressions. This project will provide one such database and thus could potentially accelerate research in this field. Automated recognition systems would have a tremendous impact on basic research by making facial expression measurement more accessible as a behavioral measure, providing data on the dynamics of facial behavior at resolutions that was previously unavailable. Such systems would also lay the foundations for computers that can understand this critical aspect of human communication.

DESCRIPTION (provided by applicant): The goal of this training program is to provide advanced graduate students and postdoctoral fellows with interdisciplinary training in cognitive neuroscience and to create a new generation of cognitive neuroscientists devoted to the study of normal and abnormal brain function. San Diego has an exceptionally active research community in cognitive neuroscience, and there are 23 participating faculty in this program from the Salk Institute for Biological Studies and at UCSD, including 8 new faculty. The graduate students will be enrolled in graduate programs in the Departments of Cognitive Science, Psychology, the Neurosciences Graduate Training Program and the Division of Biological Sciences. The predoctoral trainees will be engaged in thesis research and the postdoctoral trainees will be engaged in new research projects aimed at developing new approaches to understanding cognitive brain functions. The administrative structure of the proposed program comprises an Executive Committee of T. Sejnowski, Program Director, Salk and UCSD;T. Albright, Salk;E. Bates, UCSD;W. Bechtel, UCSD;U. Bellugi, Salk;S. Hillyard, UCSD;M. Kutas, UCSD;L. Squire, UCSD;D. Swinney, UCSD. The four major areas of research of the faculty are Vision, Memory and Attention, "Language and Development, and Sleep and Consciousness. The laboratories that will participate in the proposed training program have a long history of research collaborations. Training will be provided in a wide range of techniques including electrophysiology, psycholinguistics, functional magnetic resonance imaging, and computational modeling. In the last 5 years the research facilities have been expanded with the establishment of a major new UCSD/Salk functional MRI Center and a new Swartz Center for Computational Neuroscience at UCSD, and graduate students will also be drawn from a new Training Program in Computational Neurobiology. The current training program supports 4 predoctoral and 4 postdoctoral trainees. Given the significant increase in the number of faculty over the last 5 years, the increase in the number of qualified applicants, and the availability of major new facilities for pursuing research in cognitive neuroscience at UCSD, we are requesting support for 6 predoctoral and 6 postdoctoral trainees.

DESCRIPTION (provided by applicant): The goals of this project focus on understanding the origins and consequences of the stereotyped patterns of spike times observed in response to repeated injections of the same stimulus. These patterns have been observed in vivo in the sensory periphery in mammals and invertebrates alike, and even in a given cell type across different animals. The existence of stereotyped patterns leads to novel statistical structure of spike trains that in turn may have profound effects on the encoding and decoding of stimulus-related information (information about the world) by subsequent neurons in a given pathway. Patterns would not occur in neurons if they simply responded at random intervals according to their average firing rate. However, patterns are seen in biophysically realistic neurons in which synaptic currents, nonlinear spike-initiation and refractory mechanisms combine with the dynamics of input currents to create complex, highly structured behavior. Experiments using in vitro slices of rat frontal cortex and computer simulations of neural models will be used to study input-output relationships to a wide variety of input stimuli, to examine the abilities of neurons to code information in patterns of spikes and also to decode patterns. The conditions that occur in vivo will also be recreated experimentally in vitro using a dynamic clamp to inject background synaptic input conductances into a cell. Inputs that will be examined include constant (current step) stimuli, sinusoidal stimuli, complex quasi-periodic (multiple band-pass) stimuli similar to the local field potential or EEG patterns, observed in vivo and aperiodic stimuli. In order to quantify the reliability of patterns in spike trains, a new measure based on the correlations between spike trains will be used. A new clustering algorithm will also be used to identify and extract spike patterns from raw data. As preliminary results, we show that these new tools can extract previously unknown spike patterns from published data obtained in vivo. A better understanding of how the spike trains encode temporal information in spike timing is important in a variety of neurological diseases such as dyslexia, autism and schizophrenia, where there is evidence for a breakdown in the processing of precise timing information in the brain.

Computer algorithms that analyze auditory scenes and extract individual sound sources would have a strong impact in several domains. For example, to facilitate natural interaction with computing devices by voice, an automatic speech recognition system must be able to focus on the voice of the person speaking to it and ignore sounds from all other sources. A hearing device must perform a similar task to allow a hearing impaired person conduct a conversation in a noisy, multiple source environment. Building on recent advances in the fields of machine learning and signal processing, we are developing sophisticated adaptive algorithms for analyzing auditory scenes with multiple sound sources. Our algorithms are based on probabilistic modeling of different sound sources and of the manner in which they overlap each other and distorted by reverberation and background noise. We use advanced recent techniques for inferring our models from sound data captured by a microphone array, separating those data into individual sources, and automatically determining the type of each source present and its location. Moreover, by reconstructing the clean signal of individual sound sources, we dramatically enhance the accuracy of automatic speech recognition for human speakers in multiple source environments. To facilitate the development and evaluation of our algorithms, and also to encourage competition between other research groups ultimately resulting in improved techniques, we collect a large dataset of multiple source auditory scenes, and make it publicly available on a dedicated website.

Proposal #0847752
PI- Gert Cauwenberghs

ABSTRACT

This exploratory research project aims to observe and augment the learning experiences of children through non-intrusive acquisition, on-line analysis and interpretation of their brain dynamics. Current systems for recording high-resolution encephalogram (EEG) dynamical brain activity are not suitable for this purpose because they distract the children and constrain their mobility by excessive wiring between electrodes and computer. Existing methods are also not useful because of unreliable contact between electrodes and scalp during body motion. This two-year project specifically entails the design and implementation of a low-weight wearable, wireless EEG recording system with 128 embedded non-contact electrodes. This will include supporting software for real-time analysis and display of brain dynamics on a host computer. The research will give rise to new methods for non-intrusive acquisition and on-line interpretation of brain dynamics, and open up new research directions not possible using existing methods. The project supports inter-disciplinary graduate research combining biophysics of EEG, engineering of non-contact and wireless sensors, independent component analysis, cognitive neuroscience, and the temporal dynamics of learning.

Outcomes of this research will contribute to the broader understanding of brain function at a level combining cognitive neuroscience and social dynamics. A diverse and interdisciplinary body of students at the NSF Temporal Dynamics of Learning Center (TDLC) and the Institute of Neural Computation at UCSD will take part in applying the instrumentation to learning research in the biological, cognitive and social sciences. The wireless EEG infra-structure also significantly enhances the mobility of EEG recording in existing motion-capture facilities at TDLC, allowing the study of learning in dynamic environments with freely interacting subjects. The project will further benefit from outreach channels supported by the TDLC, including the Howard Hughes Medical Institute (HHMI) Hughes Scholar Program (HSP), and the Preuss High School at UCSD.

Address; CRISP; Chemicals; Computer Retrieval of Information on Scientific Projects Database; Diffusion; Electrons; Extracellular Space; Funding; Grant; Institution; Intercellular Space; Investigators; Modeling; NIH; National Institutes of Health; National Institutes of Health (U.S.); Negative Beta Particle; Negatrons; Nerve Impulse Transmission; Nerve Transmission; Nerve Transmitter Substances; Neuronal Transmission; Neurotransmitters; Physiology; RFP; Request for Proposals; Research; Research Personnel; Research Resources; Researchers; Resources; Services; Simulate; Site; Source; Structure; Synapses; Synaptic; United States National Institutes of Health; base; chemical release; neurotransmission; reconstruction; simulation

[unreadable] DESCRIPTION (provided by applicant): The overall goal of the proposed research is to determine how brain events at the biophysical level in the human brain influence the macroscopic recordings outside the skull. The specific focus is on sleep spindles, the best studied sleep rhythm, for which we will obtain simultaneous EEG, MEG and depth electrode recordings. These empirical data will be analyzed to determine the laminar sources of the currents within the cortex associated with spindles and these results will be matched to predictions from computational models. We will develop three types of related network models to cover a wide range of spatial scales. The working hypotheses are, first, that spindles are generated by multiple loosely-coupled cortical regions through rhythmic activation of thalamocortical afferents, onto superficial cortical layers from the matrix system, and onto middle layers from the core system, and second, that spindle oscillations in restricted thalamocortical domains in the core system are spread via the matrix system. Empirical Specific Aims and Hypotheses 1) Record MEG and EEG simultaneously during sleep spindles. Hypothesis: MEG and EEG will record very different activity patterns at the sensor level and in their inferred sources. Poor correlation will be apparent as inconstant phase and amplitude relations within spindle discharges, as well as only loose correlations as to when spindles occur. 2) Record from intracranial EEG macro-electrodes during sleep spindles, with simultaneous scalp EEG and MEG recordings. Hypothesis: Recordings from different cortical generators will be only loosely coupled with each other, or with scalp EEG, thus supporting the general view suggested by MEG. 3) Record from intracranial microelectrode arrays during sleep spindles. Hypothesis: Neuronal generator currents vary during each spindle between superficial and middle layers. 4) Reconstruct supragranular and infragranular pyramidal cells from association cortex in humans, determine their laminar distribution, and estimate the likely termination zones of core and matrix thalamocortical projections. Hypothesis: Significant anatomical differences will be found between human association cortex and the rodent sensory cortex. Modeling Specific Aims and Hypotheses 1) Construct accurate, realistic EEG/MEG forward solutions based on cortical reconstruction from structural MRI. Hypothesis: Basic parameters of extracranial EEG/MEG can be replicated by modeling multiple thalamocortical domains with varying synchrony between domains. The values of the fit parameters to EEG/MEG will match those inferred from intracranial macroelectrode recordings. 2) Analyze the recorded cortical Current Source Density (CSD) using Principal Components Analysis (PCA). Model the CSD patterns expected from the matrix and core thalamocortical afferents using reconstructions of supragranular and infragranular pyramidal cells, their population distribution, and terminations of matrix and core afferents. Hypothesis: The main spatiotemporal CSD components contributing to the spindle identified with PCA will correspond to the CSD components modeled to result from activation of matrix and core thalamocortical afferents. 3) Construct neuronal models based on Hodgkin-Huxley ionic currents that include cortical cells, thalamic reticular nuclear cells, matrix and core thalamic relay cells, and their interconnections in a minicolumn. Hypothesis: The spatiotemporal patterns of currents predicted in different cortical layers will match those obtained using the methods described in modeling aim 2. 4) Scale up the cortical minicolumn network model using simplified neural models to a spatially accurate cortical model that can generate EEG/MEG patterns. Hypothesis: The predicted EEG and MEG from the model will match those observed in recordings, with the involvement of matrix vs core thalamocortical systems corresponding to the parameters derived in modeling aim 1. 5) Develop a statistical model that matches the properties of the scaled up model and analyze it with methods from statistical physics. Hypothesis: The matrix system controls the effective cortical connectivity and the core system controls the local correlation length. Collaborative research Recordings from humans will be performed at MGH (Cash), with analysis and modeling at UCSD (Halgren and Sejnowski). These 3 teams of PIs, students and postdoctoral fellows will interact on a daily basis during the research and will meet formally at least once a year to assess progress and plan new experiments. [unreadable] [unreadable] [unreadable]

DESCRIPTION (provided by applicant): The communication between neurons at synapses occurs in small spaces with small numbers of molecules, far from equilibrium in unmixed volumes. The dynamics of chemical reactions in microdomains is difficult to estimate without having an accurate model of sub-cellular ultrastructure as well as detailed knowledge of the locations and kinetic rate constants of all the relevant molecules, including the neurotransmitter receptors, transporters, binding proteins, degradative enzymes, and other signaling targets. For example, common signaling agents such as calcium have different effects depending on where they enter the cell, and where their targets are located. If the spatial organization of the cell is important, then it is not enough to reconstruct the reaction network of signal transduction pathways. Clearly, to study and understand the behavior of these signaling pathways it is essential to obtain accurate three-dimensional (3-D) anatomical reconstructions of the pathways; that is, to place the signaling pathways within their natural context, which includes the cellular ultrastructure and 3-D distributions of the biochemical molecules. Here, using the MCell Monte Carlo computational modeling program and high-resolution 3-D reconstructions of neural tissue, we propose to explore three components of synaptic signaling: 1) calcium dynamics in the presynaptic boutons from area CA3 pyramidal cells in the rat hippocampus, 2) calcium microdomains in the vicinity of ligand-gated ion channels in postsynaptic calyciform synapses of the avian ciliary ganglion, and 3) extracellular glutamate dynamics in glomeruli from rat cerebellar cortex. These three systems are sufficiently well characterized for quantitative modeling and will allow us to explore the mechanisms underlying the release of neurotransmitter following the entry of calcium into the presynaptic terminal and cross-talk between release sites, the effects of calcium entry into the postsynaptic cell, and the diffusion of neurotransmitter in the synaptic cleft and spillover to neighboring synapses in extracellular space. The detailed level of understanding of these systems afforded by these MCell models will provide new insights that may be applicable to many other synapses, and in particular should help to elucidate how dysfunctions in signaling microdomains may contribute to neurological and psychiatric pathology. PUBLIC HEALTH RELEVANCE: Neurons communicate at synapses, where chemicals released by the presynaptic neuron bind to receptors on the postsynaptic neuron and open channels in the membrane that ions flow through. We will study every aspect of this process using a computer model, called MCell, which simulates every important molecule and chemical interaction between them during synaptic signaling. These studies will help us understand how synapses work and how they dysfunction in neurological and psychiatric pathology.

The goal of the SGER project is to develop new tools to understand the principles of gene network operation and to use these principles as the basis for a new kind of geno-mimetic network architecture. Gene networks display multiple (degenerate) ways of giving a common output. The project identifies what principles these degenerate networks have in common and incorporate such common principles into network and circuit architectures for non-biological computational devices.

The project constitutes a new approach to the question of whether there are common underlying principles to biological network operation. The project also presents the potential to introduce an entirely new class of network architecture based on the functional configurations of gene networks.

Project Summary/Abstract The goal of this research is to understand why cerebral cortical trauma often leads to paroxysmal activity. Within 24 hours following head injury, up to 80% of patients with penetrating wounds display clinical seizures. Such acute seizures often initiate epileptogenesis the subthreshold processes that lead to spontaneous, recurring seizures and ultimately to epilepsy. We propose to study the electrophysiological features of trauma- induced epileptogenesis in chronic experiments in vivo, in vitro and with computational models that will be developed in close contact with the experiments. The primary hypothesis for the cause of epileptogenesis that we will test is that trauma-related chronic blockade of activity may activate homeostatic plasticity mechanisms that upregulate depolarizing influences (such as excitatory intrinsic and synaptic conductances) and downregulate hyperpolarizing ones (such as inhibitory conductances). Under the abnormal conditions found in traumatized cortex, this may create an unstable balance that leads to paroxysmal seizures. Multisite local field potential recordings (up to 64 channels) will be used to test the hypothesis that invasive brain trauma creates heterogeneous under- and overexcited cortical areas and that interaction of these areas increases the likelihood of seizure occurrence. Direct evidence for the role of homeostatic plasticity in the epileptogenesis will be obtained by measuring changes in minis, synaptic responsiveness, axonal arborization, intrinsic cellular properties, and multisite focal field potentials. Measurement will be performed over the medium-term (days) and long-term (weeks). In vivo electrophysiological semichronic and chronic experiments, in vitro experiments from chronically deafferented cortical slices as well as morphological studies will be performed at Laval University (Canada). Data from studying the conditions that increase the likelihood of seizure development after brain trauma will be studied using Independent Component Analysis (ICA) at the Salk Institute and will be incorporated into Hodgkin-Huxley type models of cortical neurons and networks at the UC Riverside. The goal of the computational models is to explore the interplay between all of the changes that occur in the cortex in vivo during epileptogenesis and to make predictions for interventions that could prevent seizures. The design of these interventions will be based on approaches that could be further developed to treat humans with trauma-induced epilepsy in clinical settings.

DESCRIPTION (provided by applicant): The immediate objective of this proposal is to build an accurate dynamic model of activation and autophosphorylation of the signaling protein Ca2+/calmodulin (CaM)-dependent protein kinase II (CaMKII) during influx of Ca2+ into the postsynaptic spine through NMDA receptors. The work will proceed in three stages. First, investigators will validate a model of activation of individual monomeric catalytic subunits by Ca2+ and CaM and refine its kinetic parameters by comparing the model to experiments. The deterministic model is implemented in Mathematica; the output of the model will be tested against bench assays of the enzymatic activity of monomeric subunits of CaMKII under a wide range of concentrations of the subunits, CaM, and Ca2+. The concentrations will mimic both in vivo and in vitro conditions. In the second stage, investigators will construct a model of activation of the dodecameric holoenzyme of CaMKII, based on the model validated in the first stage. They will model cooperative activation of subunit dimers within the holoenzyme, and three different paths of autophosphorylation of its subunits. The models will be constructed in the program MCell, which supports spatially correct stochastic models of protein interactions and enzymatic activation, within biologically realistic geometries. This model will employ a new rule-based algorithm to specify the locations and behavior of subunits in holoenzymes. It will be constructed in a well-mixed volume to enable testing by comparison to bench experiments with holoenzymes under a wide range of concentrations of subunits, CaM, and Ca2+. The comparisons will be used to optimize four new parameters in the holoenzyme model, and to choose the most accurate model for progression of autophosphorylation within the holoenzyme. In the third stage, investigators will introduce optimized models of the CaMKII holoenzyme into a larger MCell model of Ca2+ influx into spines through NMDA receptors and its removal by pumps and exchangers. Simulations in MCell with this model will be used to test hypotheses about parameters governing activation of CaMKII in spines. The intellectual merit of the proposal lies in its utility in the study of mechanisms of learning in the central nervous system. The regulatory machinery in a spine controls synaptic strength by regulating activity-dependent changes such as LTP and LTD. We know much about the regulatory enzymes in a spine and we have hypotheses about enzymatic networks that regulate the cellular processes controlling synaptic plasticity, including insertion and removal of glutamate receptors and changes in the shape of the spine actin cytoskeleton. However, at the present stage of analysis, qualitative studies with mutant animals, or over-expression and knock-down of particular enzymes are the dominant paradigm in the field and they are not adequate to bring our knowledge to the next level, which is to establish the timing of the action of each of these players, and the precise conditions and position in the regulatory network at which each one becomes important. To reach that level of understanding, we need better quantitative models and methods. CaMKII is one of the the initial enzymes activated by Ca2+ coming through NMDA receptors during induction of LTP. A well-validated quantitative model of its activation in the powerful MCell program will provide a starting point and an example for the construction of dynamic models of successive steps in spine regulatory pathways. The broader impacts include the educational goal of fostering introduction of computational techniques into cellular neurobiological research. A female postdoctoral fellow will be trained in experimental techniques to test computational models, and in the use of MCell. Undergraduate students (including minority students) will be involved in the work through summer research programs at Caltech and Salk. All models will be made available to the community for download. The models of CaMKII holoenzymes will be a first example of simulation in MCell of interactions within a cytosolic multiprotein complex. The syntax for doing this will be published, and taught in the regular workshops on MCell sponsored by NSF. The proposal has medical significance. Deficiencies in spine signaling pathways that use CaMKII are associated with working memory deficits similar to those that underlie schizophrenia and related thought disorders. A quantitative understanding of the factors governing activation of CaMKII during synaptic activity, and its role in controlling synaptic plasticity will facilitate development of clinically useful pharmacological agents that target specific aspects of synaptic dysfunction with fewer undesirable side effects.

Intellectual Merit: This project aims at combining cognitive and computational neuroscience, neuroengineering and system identification towards a transformative understanding of the way distributed brain dynamics interact with motor activity in humans. 3-D body and limbs movement kinematics, eye movements and electroencephalographic (EEG) spatiotemporal brain data will be recorded simultaneously during motor control and adaptation in healthy and Parkinson?s disease patients. In particular, altered and real world motor tasks will be simulated in 3-D immersive virtual reality technology with force feedback robots providing proprioceptive interaction and feedback. Cognitive, behavioral and kinematics data will constrain the design of large-scale computational models of motor control and adaptation based on known anatomy and physiology of the basal ganglia. Neuromorphic engineering will guide the design of mobile embedded computational systems for real-time emulation of the brain-body models and closed-loop sensory-motor control for Parkinson?s patients. We expect that the development of new machines for neuro-rehabilitation will result in a threefold synergetic interaction between engineering and neuroscience: human-machine interactions will transform the notion of movement control and provide new contexts to study embodied cognition that will benefit neuroscience; in turn, new knowledge in neuroscience and motor control will accelerate the development of adaptive machines for rehabilitation and/or enhancement. Finally, comprehensive and predictive mathematical models of motor control implemented in neuromorphic hardware are expected to lead to new intelligent neuroprosthetic tools.

Broader Impact: Outcomes of this research will contribute to the system-level understanding of humanmachine interactions and motor learning and control in real world environments for humans, and will lead to the development of a new generation of wireless brain and body activity sensors and adaptive prosthetics devices. This will advance our knowledge of human-machine interactions, stimulate the engineering of new brain/body sensors and actuators, and have a direct influence in diverse areas where humans are coupled with machines, such as brain-machine interfaces, prosthetics and telemanipulation. We anticipate that the confluence of cognitive and computational neuroscience, control theory and wearable, unobtrusive bioinstrumentation will provide novel non-invasive approaches or the treatment and neuro-rehabilitation of Parkinson?s disease and will potentially transform our understanding of brain/body interactions. The project draws graduate and undergraduate students across divisions and in the NSF Temporal Dynamics of Learning Center (TDLC) and Institute of Neural Computation (INC) at UCSD participating in interdisciplinary engineering and neuroscience aspects of the design, optimization, and training of largescale neuromorphic systems and their human interfaces. Through outreach channels on campus supported by the TDLC and the NSF Research Experience for Undergraduates (REU), the program will actively pursue increased participation in research and education of the next generation of scientists and engineers.

It is commonly accepted that there is a crisis in education in the US. There are too many struggling learners, too many students who cannot read or do basic arithmetic, let alone advanced mathematics. What is not commonly accepted is what to do about this crisis. The researchers at the Temporal Dynamics of Learning Center (TDLC) believe that part of the current crisis in education is the lack of scientific understanding of how the brain learns, and the lack of translation of this scientific understanding to the classroom. An essential, yet understudied, component of learning that could have a strong impact on education is the role of time and timing in learning. TDLC brought together an interdisciplinary team of over 40 investigators from 16 different research institutions in order to focus research energy on this goal. TDLC's purpose is to achieve an integrated understanding of the role of time and timing in learning, across multiple time and spatial scales, brain systems, and social systems, to 1) create a new science of the temporal dynamics of learning; 2) to use this understanding to transform educational practice; and 3) to create a new collaborative research structure, the network of research networks, to transform the practice of science.

Why study timing? Timing is critical for learning at every level, from learning the precise temporal patterns of speech sounds, to learning when to give feedback in the classroom, to the optimal frequency and timing of studying new material. Moreover, a decade of neuroscience research demonstrates that the intrinsic temporal dynamics of the brain itself also reinforce and constrain learning. For example, work at TDLC has shown that measurements of the brain waves of a toddler-the temporal dynamics of thought - can predict how well that child will perform at language tasks years later. This provides the possibility that early intervention could overcome these difficulties, demonstrating the usefulness of studying temporal dynamics. A research program of this size and scope is clearly only possible through the Center Funding model, in order to provide resources at the scale necessary to coordinate the large team of researchers. The work is organized by dividing the personnel into four research networks, where researchers from multiple disciplines are interested in common questions, and who synchronize their research around experiments that can be carried out in humans, animals, and computational models, allowing unprecedented convergence of techniques on a single question.

This EAGER award funds a theoretical study of a proposed new technology for measuring neural currents within brain tissue. This research project is headed by Professors Lawrence Frank and Terrence Sejnowski at UC San Diego.

The research program funded by this grant concerns new methods of understanding and measuring the electrical processes that occur within the human brain and which signal brain activity. Current techniques involve functional magnetic resonance imaging (FMRI), but these methods only operate indirectly by triggering on the electrical changes that occur in the blood surrounding the brain or within the brain. Such techniques are therefore limited by the fact that blood flow induces a certain fundamental inaccuracy having to do with spatial separation (the distance of blood flowing prior to FMRI detection) as well as a time lag which separates the time of the original electrical activity in the brain and its subsequent impact on the surrounding blood. In this study, by contrast, the PI's will analyze a new technology which might be able to completely overcome these limitations by focusing directly on the neural currents within the brain tissue itself. This new technology is referred to as "neural current MRI", and the PI's will model the effects of these neural currents and the possibilities for their detection by conducting a detailed numerical simulation of such brain activity.

The implications and broader impacts of such a project are significant. The results of such a study might help to pave the way for a whole new technology by which we might be able to image the brain with a precision never previously achieved. Moreover, because the research in this proposal crosses many traditional boundaries (including biology, physics, and computational mathematics), it is highly interdisciplinary. The results of this study therefore have the potential to benefit research in a variety of related disciplines.

The brain is composed of two major cell types: Neurons and glial cells. Glial cells are traditionally regarded as the brain's supportive cells. However, many lines of work over the past decade have documented that glial cells may also participate in complex neural processes and thereby comprise an integral element of higher cognitive function, such as working memory, learning, and sleep. Other lines of work have shown that human astrocytes are larger and structurally more complex than astrocytes in the rodent brain. In support of this concept, transplantation of human glial cells into mice resulted in generation of mice that were faster learners and performed better on memory tests. However, existing computational modeling techniques employed for understanding the processes involved in learning and memory do not include glial cells. The aim of the proposed research is to: 1) Develop computational modeling techniques that incorporate glial cells. 2) Use these novel computational modeling techniques to make predictions regarding the role of glial cells in learning and memory. 3) Test the predictions using a combination of patch clamping and Ca2+ imaging. 4) Use the data collected to continuously refine the computational modeling techniques.
The broader impact of this proposal will be to further the scientific understanding of underappreciated, yet essential substrates of learning and memory. Including glial cells in addition to neurons in modeling approaches additionally carries the hope of increasing computational power and processing capabilities of adaptive learning technology, in addition to improving the performance of bio-integrated prostheses for individuals with impaired learning or other debilitating neurological disorders.

DESCRIPTION (provided by applicant): Neural oscillations organize cortico-thalamic activity, and their role in memory, attention and sleep are a central focus of systems neuroscience. Sleep spindles are among the most prominent oscillations, and have been studied at many levels of investigation, from the biophysical level, where the low threshold calcium currents are implicated in the waxing-and-waning 11-15 Hz bursts of spikes that originate in the thalamus and recruit cortical circuits, to the systems level where the electroencephalogram (EEG) and magnetoencephalogram (MEG) measured outside the skull register largescale spatial and temporal coherence in the bursting pattern across the cortex (Destexhe and Sejnowski, 2001). Despite the wealth of physiological, anatomical and computational studies, major questions remain to be resolved: How do nearby parts of the cortex become synchronized during spindles? How are spindles propagated across the cortex? Why is there a discrepancy between the temporal patterns of spindles simultaneously observed in EEG and MEG measurements? What are the consequences of spindle activity in thalamocortical systems for cortical reorganization and memory consolidation during sleep? We propose to attack these questions with a range of experimental and modeling techniques that 1) link detailed models at the biophysical level to recordings from humans at the level of current source density analysis (CSD) recordings from depth electrodes; and 2) relate large scale reduced models of cortical circuits to EEG and MEG measurements in humans. This is the first time that all of these powerful empirical and modeling approaches have been integrated into a single, multiscale approach to understanding the origin of macroscopic field measurements outside the scalp based on the specific biophysical mechanisms occurring in neurons located in different layers of the cortex and thalamus.

The global acceleration in technological innovation and transformation is incidentally leading to a scientific culture in which technology is often designed and launched without providing ethical guidelines for its use. Thus, the mere pace of technology mandates an urgent need for establishing ethical guidelines as part of the natural course of science. Ultimately, a sustainable landscape of innovation will include a culture of partnership between the scientific community and ethicists, allowing capitalization on the benefits of discovery while mitigating the liabilities to society. This movement must be driven as a team effort at the outset. Thus, a primary purpose of this proposal is to formulate a cross-disciplinary team of scientists and ethicists to consider the ethical use of a select set of emerging technologies for application to the science of learning, education, rehabilitation, medicine, and augmented humans.

At the outset, a team of engineers, cognitive scientists, psychologists and educators will work alongside ethicists with expertise in the ethics of virtual reality, neurorehabilitation, robotics, wearable sensors, and augmented humans through a workshop-style forum, set at the University of Queensland, Australia, in order to forge a path for establishing ethical guidelines as part of the scientific process. A virtual organization will be formed to sustain these efforts and to create a secure forum for continual interaction between scientists and relevant ethicists and policy makers. The international component of this grant is absolutely essential not only to the training of a diversity of students who will be trained to lead sustainable science in the future, but also to balance the venture as an international problem worthy of coordination and international collaboration at the outset. Both the international component and the virtual forum also establish the fact that ethical guidelines and policy need to be an inherent component of STEM education. Educational activities for graduate students are an inherent component of each goal of the grant and programs for recruitment and inclusion of under-represented students are embedded in the plan. Additional STEM education materials, and established routes for broader dissemination to under-represented students, will be produced as a byproduct of the high level scientific program. The activities will conclude with a fully articulated process for establishing ethical guidelines as an integral part of technology development.

The success of the BRAIN initiative will depend on widespread access to the technological advancements, computational tools, and data sets created by the initiative. However, there are no existing mechanisms for providing national access to the increasingly technologically and computationally oriented investigations of the brain. The barriers to entry are both financial and structural: not only is technologically intensive neuroscience costly, it requires an investment in physics, engineering and computer science beyond the scope of individual laboratories. This prevents the community's efficient utilization of current technological capabilities and limits the questions and hypotheses that will drive the next generation of innovation. Thus there is a need to counteract the widening gap between the small fraction of laboratories developing and utilizing the most recent technology and the remaining majority of neuroscientists. The successful removal of the gap will require a sophisticated national clearing house to ensure that the correct physics, engineering, and computer science tools are vetted and freely accessible for measurements of brain structure and functions. Successful accomplishment of these goals will require an iterative process whereby specific needs of the neuroscience community will be identified and either paired with the appropriate scientific, technological and computational resources or pipelined for potential future innovation. The model for the operation of this project will be a user facility, housed at Argonne National Laboratory (ANL), and leveraging the existing resources of their science facilities. This award provides funding for seed grants for infrastructure development, conferences, education, and outreach.

The team will enlist the Physics of Living Systems community, most specifically the young scientists therein, to join the neuroscience research effort by connecting to the graduate research network led by the NSF Physics Frontier Center for Theoretical Biological Physics. In order to engage and train a broad community, several annual conferences will be held that will cover a broad range of topics in imaging and quantitative neuroscience. The team will augment the program run by the UC Neuroscience Institute to teach Neuroscience to local 7th/4th graders. Almost all of the students in the target schools are African American and live in the local South Side community. ANL will partner with this endeavor by support through its own educational programs, but for the first time broaching the technology of neuroscience.

This project will integrate the capabilities of deep learning networks into a biologically inspired architecture for sensorimotor control that can be used to design more robust platforms for complex engineered systems. Studies of the flexibility and contextual aspects of sensorimotor planning and control will extend existing paradigms for human-robot interactions and serve as the foundation for creating personal assistants that are able to operate in natural settings. Growing understanding of how these layered architectures are organized in the brain to produce highly robust, flexible, and efficient behavior will have many applications to rapidly evolving technologies in complex environments, including the Internet of Things, autonomous transportation, and sustainable energy networks.

The nervous system is a layered architecture, seamlessly integrating high-level planning with fast lower level sensing, reflex, and action in a way that this project aims to both understand more deeply and mimic in advanced technology. The central goal is to develop a theoretical framework for layered architectures that takes into account both system level functional requirements and hardware constraints. There are both striking commonalities and significant differences between biology and technology in using layered architectures for active feedback control. The most salient and universal hardware constraints are tradeoffs between speed, accuracy, and costs (to build, operate, and maintain), and successful architectures cleverly combine diverse components to create systems that are both fast and accurate, despite being built from parts that are not. Recent progress has made it possible to integrate realistic features and constraints for sensorimotor coordination in a coherent and rigorous way using worst-case L-infinity bounded uncertainty models from robust control, but much remains to explore to realize its potential in neuroscience and neuro-inspired engineering. Another application of this framework will be to software defined networking, which explicitly separates data forwarding and data control, and provides an interface through which network applications (such as traffic engineering, congestion control and caching) can programmatically control the network. This makes it a potential "killer app" for a theory of integrated planning/reflex layering; research collaborators will be eager to deploy new protocols on testbeds and at scale.

IV. TR&D2 - Abstract The overall goal of this project is to develop the next generation computer simulation platform for spatially realistic simulation and analysis of cellular and subcellular biochemistry. Cellular systems, especially in neurons, are profoundly difficult to understand because of the interplay between spatial, biochemical and molecular complexity that occurs on multiple levels of organization, from macromolecular assemblies to synapse architecture to neural circuits. Biological complexity is daunting and scientific investigators must persevere to finds ways to overcome it. This is important because Scientific Discovery is driven by testable hypotheses which derive from our intuition and questions surrounding our current understanding of reality. But when daunting complexity confounds our intuition we struggle to conceive new hypotheses and the cycle of discovery grinds to a halt. Computational models allow investigators to probe the complex relationships between biological components, obtain new insights and intuition -- the genesis of new hypotheses. The MCell/CellBlender platform for cell modeling we are developing is expressly designed to fulfill this need, providing insight and understanding of complex cellular systems. The cell modeling tools we develop here are designed to mesh with the molecular, network, and image-derived modeling tools of TR&Ds 1, 3 and 4. The tools will be used by our Driving Biomedical Project research partners to study neuronal and synaptic structure and function and the intricate biochemical pathways involved in learning and memory in the brain. The detailed level of understanding of these systems afforded by computational modeling of these systems will provide new insights that may be applicable to many types of cell signaling pathways, and in particular should help to elucidate how dysfunctions in cell signaling may contribute to neurological and psychiatric pathology.

The overall goal of this project is in the spirit of Richard Feynman -- What I cannot create, I do not understand. Over the past decade many laboratories have reported new insights on synaptic transmission properties, based on electrophysiology, imaging, or biochemistry experiments. But the precise impact on synaptic properties of the various measured parameters such as molecular organization, mobility, or molecular interactions has remained elusive. Here we propose to overcome this problem through a synergy between the phenomenological descriptions of AMPAR organization based on super-resolution imaging techniques, the identification of the core of such organization with quantitative biochemistry, and quantitative biophysical modeling of synaptic transmission. The resulting model will deliver testable predictions of synaptic transmission properties that we will directly test with electrophysiological recordings. The ultimate result will be a reliable model based on brand new knowledge of synaptic organization and function at the nanoscale. The project will be divided into three Aims. In Aim 1 we will implement in an existing model three main modifications regarding newly available knowledge on AMPAR properties: (i) in situ constraints for the activation/inactivation kinetic rate constants of the tetrameric concerted opening model of AMPAR including the effect of AMPAR/TARP interaction on AMPAR gating properties; (ii) the tight organization of AMPAR in nanoclusters as reported recently by using super-resolution techniques; (iii) The lateral diffusion of AMPAR which has been measured with live single particle techniques. In Aim 2 we will measure the biochemical on/off rate and cross-affinity of the three main proteins responsible for AMPAR organization: PSD95, the TARPs and synGAP. At the end of this Aim, the model should be able to perfectly simulate synaptic transmission recorded at a neuron, both in term of kinetics, amplitude and variability, by taking into account the lateral mobility of AMPAR complex and PSD95 slot occupancy as a function of the determined affinity. Finally in Aim 3, we wfll validate the model and the hypothesis by modifying either the expression level of synGAP or the relative affinity of synGAP or TARP for PSD95, and then compare effect of such changes on AM PAR organization and dynamic properties and on synaptic transmission properties with the model predictions. RELEVANCE (See instructions): The proposed work involves study of the molecular mechanisms that control synaptic plasticity and their role in mental illness, a Strategic Research Priority of the NIMH. The work will help to clarify the function of the protein synGAP in CNS synapses and will impact Public Health as SynGAP has been found to be mutated in -1 % of children presenting a cognitive disability accompanied by autism and or epilepsy.

Abstract The goal of this research is to develop new multivariate data analysis techniques for neural recordings that reveal causal dependencies between recording sites. Delay Differential Analysis (DDA) is a robust and ef?cient nonlinear time-domain algorithm for time series data that complements linear spectral methods. DDA combines delay and differential embeddings in nonlinear dynamical systems to discriminate between different normal and abnormal cortical states with high temporal resolution and insensitivity to artifacts. The proposed research generalizes Granger causality for linear systems by developing a cross-dynamical version of DDA (CD-DDA) to measure the ?ow of information between brain areas. This is an important problem for which existing approaches are inadequate. CD-DDA will be applied ?rst to simulations of cortical network models with Hodgkin-Huxley neurons, where causal in?uence can be controlled and the ef?cacy of CD-DDA can be validated. In collaboration with Sydney Cash at the Massachusetts General Hospital, CD-DDA will then be applied to electrocorticography (ECoG) recordings from human epilepsy patients with implanted grids of electrodes. We previously analyzed these recordings with DDA, which revealed differences between cortical states leading up to seizures, abrupt shifts at the onsets of the seizures and altered cortical states long after the seizures. These ECoG recordings will be re-analyzed using CD-DDA, which should reveal how communication between cortical areas recon?gures before seizures. We also have access to many hours of interictal recordings, which will give us the opportunity to establish a baseline for how information ?ows in cortical circuits during more normal cortical activity. We will make the software for all of the DDA algorithms we have developed openly available. These new algorithms will have many other applications for analyzing neural signals online in other brain areas and from other neural time series, including calcium ?uorescence imaging from single cells, dendrites and synapses and recordings using voltage-sensitive dyes.

Project Summary/Abstract The goal of this research is to understand why cerebral cortical trauma often leads to seizures and to propose interventions that may reduce or prevent trauma-induced epileptogenesis. Within 24 hours following head injury, up to 80% of patients with penetrating wounds display clinical seizures. Such acute seizures often initiate epileptogenesis?the subthreshold processes that lead to spontaneous, recurring seizures and ultimately to epilepsy. The primary hypotheses of this project are: 1) Trauma-related chronic blockade of activity activates homeostatic plasticity mechanisms that upregulate depolarizing influences (such as excitatory intrinsic and synaptic conductances) and downregulate hyperpolarizing ones (such as inhibitory conductances); in traumatized cortex, this may create an unstable balance of excitation and inhibition that leads to paroxysmal seizures; 2) The effect of the pathological homeostatic changes is age dependent with older animals being more prone to seizures; 3) External interventions designed to prevent decrease of activity after trauma reduce the likelihood of epileptic seizures. Importantly, rather than focus on the ways to treat epilepsies after epileptogenesis is complete, this proposal aims to develop new techniques that can interfere with a process of epileptogenesis itself. Following past experiments with cats in the Timofeev laboratory, a well-established undercut model of cortical deafferenation will be used to induce seizures in mice experiments in vivo and in vitro. Measurement will be performed over the medium-term (days) and long-term (weeks). Interventions will be explored that can prevent epileptogenesis using pharmocogenetic stimulation to block homeostatic changes. In vivo electrophysiological semichronic and chronic experiments will be performed at Laval University (Canada). In vitro experiments from deafferented cortical slices will be conducted at Laval University and UCSD. Necessary data on the astrocyte properties will be provided by the collaborators (Dr. Nedergaard, Univ of Rochester). Experimental data will be analyzed at The Salk Institute and UCSD and will be incorporated into large-scale network models of the neocortex, implementing subcellular, circuit and network level properties, at the Salk Institute and UCSD. The computational models allow the interplay between all of the changes that occur in the cortex in vivo during epileptogenesis to be simulated to identify the critical mechanisms and to make predictions for intervention strategies that could prevent epileptogenesis.

Fast glutamatergic synaptic transmission is based on a precise and complex molecular organization which requires the control of the number of AMPA-type glutamate receptors (AMPARs) at the postsynaptic sites of glutamatergic synapses on dendritic spines. The number of AMPARs varies as a function of pre- and postsynaptic activation history of the synapse. It is now well described that synapses can change their number of AMPARs and therefore, their response properties through biochemical mechanisms of synaptic plasticity. In this way, information is stored in the brain. The overall goal of this project is to use quantitative models and experiments to answer two fundamental questions about the role of an abundant postsynaptic protein, synGAP, in regulation of the numbers of AMPARs. Numerous experiments in intact neurons have revealed that the level of synGAP expressed at synapses is inversely correlated with the amount of AMPARs available at the synapses, and that synGAP helps to regulate changes in AMPAR numbers during synaptic plasticity. The enzymatic GAP domain of synGAP acts as a ratchet to adjust the rates of addition and removal of AMPARs from the surface of the dendrite. SynGAP also contains a sight that binds tightly to the major scaffold protein PSD-95 via its three protein-binding PDZ domains. Important to the mental health mission of the NIMH, SynGAP plays a critical role in learning and memory in the Brain and mutation of SynGAP is implicated in cognitive disabilities. The project is divided into two broad Aims. In Aim 1, we will answer the question: What are the mechanisms by which synGAP controls the amount of AMPA receptor in the postsynaptic density (PSD) - by control of surface amount and/or by control of availability of PDZ domain binding sites in the synapse? We will improve our existing computational model of the competition between synGAP and AMPARs for binding to PSD-95 by incorporating it into our model of AMPAR trafficking. We will use genetics and sophisticated molecular engineering to experimentally disentangle the two mechanisms. Effects on the nano-organization of AMPARs will be measured by super- resolution fluorescence microscopy and electrophysiology. Results of these experiments will be used to constrain our model of AMPAR trafficking. Aim 2, Through the synergy of experimental and computational approaches, we will address the questions: How does the formation of the condensate between synGAP and PSD-95, and the presence of additional PDZ domain-binding proteins (GluN2 receptor subunits, neuroligin, nNOS, CRIPT, etc.) influence the nano-organization of AMPAR-TARPs in the PSD in the basal state and during synaptic plasticity? RELEVANCE (See instructions): We propose a combination of computational and experimental work that will help clarify the role of synGAP in regulation of AMPARs in CNS synapses, including its role in mental illness. The work will impact a specific medical condition termed ?SynGAP haploinsufficiency? or ?MRD5?, in which SynGAP is mutated in ~1% of children with sporadic non-syndromic cognitive disability accompanied by autism and/or epilepsy. The medical impacts of this work are potentially quite significant as it could help to point toward specific molecular interventions with therapeutics that could improve the lives of patients with these afflictions.

Successful learning and long-term memory retention are central to a successful society, starting with early education in schools and extending throughout life. For over 100 years research has shown that spaced learning is much more effective than massed learning for long-term memories. The efficiency of focused learning falls after an hour, which is paralleled in lab experiments at the level of single synapse between neurons, whose strength is saturated by focused stimulation. This project seeks to understand the synaptic mechanisms that eventually lead to continued synaptic growth on the time scale of many hours. The project hypothesis is that over this time period, regions inside the synapse open up to make room for a larger and stronger synapse. This research is the first step toward helping those with learning disabilities and new ways to enhance learning in others.<br/><br/>The goal of this research is to build imaging, analytical, and computational tools to investigate the structure of nanodomains within the synapse. The nanodomains comprise nascent and active zones of synapses. The nascent zones have a fully defined postsynaptic region but lack presynaptic vesicles and hence are silent. New EM tomographic imaging combined with new computational analyses will refine understanding of nascent zones as they recruit presynaptic vesicles and are thus converted to active zones in support of synaptic plasticity that underlies the advantage of spaced learning. Existing and newly acquired large data sets will be analyzed at scale with artificial intelligence. This research will be transformative for Data-Intensive Neuroscience and Cognitive Science. The data sets and AI tools will be shared broadly with the neuroscience community through the NSF-funded 3Dem Portal (3dem.org) at the Texas Advanced Computing Center (TACC). The objectives of this project are: 1) Create computational tools to delineate nascent zones automatically by mapping presynaptic vesicle docking sites in serial sections of synapses in the hippocampal CA1 region and dentate gyrus at various times after induction of LTP or cLTD, compared to control stimulation. 2) Apply a new computational analysis based on information theory and overall synapse size to measure the storage capacity of synapses, refining the definition of synaptic weight as encompassing the enlarged active zones obtained during the conversion of nascent zones. 3) Perform realistic Monte Carlo reaction-diffusion simulations of synaptic nanodomain 3D structure and function using MCell to provide a functional estimate for the boundary between nascent and active zones and determine how changes in nascent and active zones alter efficacy at synapses during saturation and recovery of LTP and cLTD.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.