Search

A primary goal of neuroscience is to understand how nervous systems, or assemblies of neural circuits, generate and control behavior. Testing and refining our theories of neural control would be greatly facilitated if we could reliably simulate an entire nervous system so we could replicate the brain dynamics in response to any stimuli and different contexts. More fundamentally, reconstructing or modeling a system is an important milestone in understanding it, and so, simulating an entire nervous system is in itself one of the goals, indeed dreams, of systems neuroscience. To do so requires us to identify how each neuron's output depends on its inputs, within some nervous system. This deconstruction, understanding function from input-output pairs, falls into the realm of reverse engineering. Current efforts at reverse engineering the brain focus on the mammalian nervous system, but these brains are complex, allowing only recordings of tiny subsystems. Here we argue that the time is ripe to embark on a concerted effort to reverse engineer a smaller system and that the nematode C. elegans is the ideal candidate system. In particular, the established and growing toolkit of optophysiology techniques can non-invasively capture and control each neuron's activity and scale to hundreds of thousands of experiments, across a large population of animals. Data across populations and behaviors can be combined because across individuals neuronal identities are largely conserved in form and function. Modern machine-learning-based model training should then enable a simulation of C. elegans' impressive breadth of brain states and behaviors. The ability to reverse engineer an entire nervous system will benefit systems neuroscience as well as the design of artificial intelligence systems, enabling fundamental insights as well as new approaches for investigations of progressively larger nervous systems., Comment: 23 pages, 2 figures, opinion paper

The classical Hodgkin-Huxley (HH) point-neuron model of action potential generation is four-dimensional. It consists of four ordinary differential equations describing the dynamics of the membrane potential and three gating variables associated to a transient sodium and a delayed-rectifier potassium ionic currents. Conductance-based models of HH type are higher-dimensional extensions of the classical HH model. They include a number of supplementary state variables associated with other ionic current types, and are able to describe additional phenomena such as sub-threshold oscillations, mixed-mode oscillations (subthreshold oscillations interspersed with spikes), clustering and bursting. In this manuscript we discuss biophysically plausible and phenomenological reduced models that preserve the biophysical and/or dynamic description of models of HH type and the ability to produce complex phenomena, but the number of effective dimensions (state variables) is lower. We describe several representative models. We also describe systematic and heuristic methods of deriving reduced models from models of HH type.

Neuroscience is undergoing dramatic progress because of the vast data streams derived from the new technologies product of the BRAIN initiative and other enterprises. As any other scientific field, neuroscience benefits from having clear definitions of its theoretical components and their interactions. This allows generating theories that integrate knowledge, provide mechanistic insights, and predict results under new experimental conditions. However, theoretical neuroscience is a heterogeneous field that has not yet agreed on how to build theories or whether it is desirable to have an overarching theory or whether theories are simply tools to understand the brain. Here we advocate for the need of developing theoretical frameworks as a basis of generating common theoretical structures. We enumerate the elements of theoretical frameworks we deem necessary for any theory in neuroscience. In particular, we address the notions of paradigms, models, and scales of organizations. We then identify areas with pressing needs to develop brain theories: integration of statistical and dynamic approaches; multi-scale integration; coding; and interpretability in the context of Artificial Intelligence. We also point out that future theoretical frameworks would benefit from the incorporation of the principles of Evolution as a fundamental structure rather than purely mathematical or engineering principles. Rather than providing definite answers, the objective of this paper is to serve as an initial and succinct presentation of these topics to encourage discussion and further in depth development of each topic.

We organized a workshop on the "Present and Future Frameworks of Theoretical Neuroscience", with the support of the National Science Foundation. The objective was to identify the challenges and strategies that this field will need to tackle in order to incorporate vast and multi-scale streams of experimental data from the technologies developed by the BRAIN initiative. The participants, divided in workgroups, identified five key areas that, while not exhaustive, cover multiple aspects of current challenges needed to be developed: Dynamics-statistics; multi-scale integration; coding; brain-body integration; and structure of neuroscience theories. While each area is different, there were coincidences on finding theoretical paths to incorporate biophysics, energetics, and ethology with more abstract coding and computational approaches. Each workgroup has continued to work after the meeting to develop the ideas seeded there, which are started to being published. Here, we provide a perspective of the discussions of each workgroup that point to building on the present foundations of theoretical neuroscience and extend them by incorporating multi-scale information with the objective of providing mechanistic insights into the nervous system., Comment: Workshop outcomes, 9 pages

In recent years, the field of neuroscience has gone through rapid experimental advances and a significant increase in the use of quantitative and computational methods. This growth has created a need for clearer analyses of the theory and modeling approaches used in the field. This issue is particularly complex in neuroscience because the field studies phenomena across a wide range of scales and often requires consideration of these phenomena at varying degrees of abstraction, from precise biophysical interactions to the computations they implement. We argue that a pragmatic perspective of science, in which descriptive, mechanistic, and normative approaches each play a distinct role in defining and bridging levels of abstraction will facilitate neuroscientific practice. This analysis leads to methodological suggestions, including selecting a level of abstraction that is appropriate for a given problem, identifying transfer functions to connect models and data, and the use of models themselves as a form of experiment.

The conventional impedance profile of a neuron can identify the presence of resonance and other properties of the neuronal response to oscillatory inputs, such as nonlinear response amplifications, but it cannot distinguish other nonlinear properties such as asymmetries in the shape of the voltage response envelope. Experimental observations have shown that the response of neurons to oscillatory inputs preferentially enhances either the upper or lower part of the voltage envelope in different frequency bands. These asymmetric voltage responses arise in a neuron model when it is submitted to high enough amplitude oscillatory currents of variable frequencies. We show how the nonlinearities associated to different ionic currents or present in the model as captured by its voltage equation lead to asymmetrical response and how high amplitude oscillatory currents emphasize this response. We propose a geometrical explanation for the phenomenon where asymmetries result not only from nonlinearities in their activation curves but also from nonlinearites captured by the nullclines in the phase-plane diagram and from the system's time-scale separation. In addition, we identify an unexpected frequency-dependent pattern which develops in the gating variables of these currents and is a product of strong nonlinearities in the system as we show by controlling such behavior by manipulating the activation curve parameters. The results reported in this paper shed light on the ionic mechanisms by which brain embedded neurons process oscillatory information., Comment: 15 pages, 15 figures

Many neuron types exhibit preferred frequency responses in their voltage amplitude (resonance) or phase shift to subthreshold oscillatory currents, but the effect of biophysical parameters on these properties is not well understood. We propose a general framework to analyze the role of different ionic currents and their interactions in shaping the properties of impedance amplitude and phase in linearized biophysical models and demonstrate this approach in a two-dimensional linear model with two effective conductances gL and g1. We compute the key attributes of impedance and phase (resonance frequency and amplitude, zero-phase frequency, selectivity, etc.) in the gL-g1 parameter space. Using these attribute diagrams we identify two basic mechanisms for the generation of resonance: an increase in the resonance amplitude as g1 increases while the overall impedance is decreased, and an increase in the maximal impedance, without any change in the input resistance, as the ionic current time constant increases. We use the attribute diagrams to analyze resonance and phase of the linearizations of two biophysical models that include resonant (Ih or slow potassium) and amplifying currents (persistent sodium). In the absence of amplifying currents, the two models behave similarly as the conductances of the resonant currents is increased whereas, with the amplifying current present, the two models have qualitatively opposite responses. This work provides a general method for decoding the effect of biophysical parameters on linear membrane resonance and phase by tracking trajectories, parametrized by the relevant biophysical parameter, in pre-constructed attribute diagrams.

We analyze a biophysical model of a neuron from the entorhinal cortex that includes persistent sodium and slow potassium as non-standard currents using reduction of dimension and dynamical systems techniques to determine the mechanisms for the generation of mixed-mode oscillations. We have found that the standard spiking currents (sodium and potassium) play a critical role in the analysis of the interspike interval. To study the mixed-mode oscillations, the six dimensional model has been reduced to a three dimensional model for the subthreshold regime. Additional transformations and a truncation have led to a simplified model system with three timescales that retains many properties of the original equations, and we employ this system to elucidate the underlying structure and explain a novel mechanism for the generation of mixed-mode oscillations based on the canard phenomenon. In particular, we prove the existence of a special solution, a singular primary canard, that serves as a transition between mixed-mode oscillations and spiking in the singular limit by employing appropriate rescalings, center manifold reductions, and energy arguments. Additionally, we conjecture that the singular canard solution is the limit of a family of canards and provide numerical evidence for the conjecture., Comment: 46 pages, 17 figures

The physical and chemical properties of metal nanoparticles differ significantly from those of free metal atoms as well as from the properties of bulk metals, and therefore, they may be viewed as a transition regime between the two physical states. Within this nanosize regime, there is a wide fluctuation of properties, particularly chemical reactivity, as a function of the size, geometry, and electronic state of the metal nanoparticles. In recent years, great advancements have been made in the attempts to control and manipulate the growth of metal particles to pre-specified dimensions. One of the main synthetic methods utilized in this endeavor, is the capping of the growing clusters with a variety of molecules, e.g. polymers. In this paper we attempt to model such a process and show the relationship between the concentration of the polymer present in the system and the final metal particle size obtained. The theoretical behavior which we obtained is compared with experimental results for the cobalt-polystyrene system., Comment: submitted for publication

We propose an indirect approach to the generation of a two-dimensional quasiperiodic (QP) pattern in convection and similar nonlinear dissipative systems where a direct generation of stable uniform QP planforms is not possible. An {\it eightfold} QP pattern can be created as a broad transient layer between two domains filled by square cells (SC) oriented under the angle of 45 degrees relative to each other. A simplest particular type of the transient layer is considered in detail. The structure of the pattern is described in terms of a system of coupled real Ginzburg-Landau (GL) equations, which are solved by means of combined numerical and analytical methods. It is found that the transient ``quasicrystallic'' pattern exists exactly in a parametric region in which the uniform SC pattern is stable. In fact, the transient layer consists of two different sublayers, with a narrow additional one between them. The width of one sublayer (which locally looks like the eightfold QP pattern) is large, while the other sublayer (that seems like a pattern having a quasiperiodicity only in one spatial direction) has a width $\sim 1$. Similarly, a broad stripe of a {\it % twelvefold} QP pattern can be generated as a transient region between two domains of hexagonal cells oriented at the angle of 30 degrees., Comment: To be published in Physica Scripta

We study the evolution of fronts in a bistable reaction-diffusion system when the nonlinear reaction term is spatially non-homogeneous. This equation has been used to model wave propagation in various biological systems. Extending previous works on homogeneous reaction terms, we derive asymptotically an equation governing the front motion, which is strongly nonlinear and, for the two-dimensional case, generalizes the classical mean curvature flow equation. We study the motion of one- and two- dimensional fronts, finding that the non-homogeneity acts as a "potential function" for the motion of the front; i.e., there is wave propagation failure and the steady state solution depends on the structure of the function describing the non-homogeneity.

We study the evolution of fronts in the Klein-Gordon equation when the nonlinear term is non-homogeneous. Extending previous works on homogeneous nonlinear terms, we describe the derivation of an equation governing the front motion, which is strongly nonlinear, and, for the two-dimensional case, generalizes the damped Born-Infeld equation. We study the motion of one- and two-dimensional fronts, finding that the dynamics is richer than in the homogeneous reaction term case.

We consider the Allen-Cahn equations with memory (a partial integro-differential convolution equation). The prototype kernels are exponentially decreasing functions of time and they reduce the integrodifferential equation to a hyperbolic one, the damped Klein-Gordon equation. By means of a formal asymptotic analysis we show that to the leading order and under suitable assumptions on the kernels, the integro-differential equation behave like a hyperbolic partial differential equation obtained by considering prototype kernels: the evolution of fronts is governed by the extended, damped Born-Infeld equation. We also apply our method to a system of partial integro-differential equations which generalize the classical phase field equations with a non-conserved order parameter and describe the process of phase transitions where memory effects are present.