A GPU-based computational framework that bridges neuron simulation and artificial intelligence.

Biophysically detailed multi-compartment models are powerful tools to explore computational principles of the brain and also serve as a theoretical framework to generate algorithms for artificial intelligence (AI) systems. However, the expensive computational cost severely limits the applications in both the neuroscience and AI fields. The major bottleneck during simulating detailed compartment models is the ability of a simulator to solve large systems of linear equations. Here, we present a novel Dendritic Hierarchical Scheduling (DHS) method to markedly accelerate such a process. We theoretically prove that the DHS implementation is computationally optimal and accurate. This GPU-based method performs with 2-3 orders of magnitude higher speed than that of the classic serial Hines method in the conventional CPU platform. We build a DeepDendrite framework, which integrates the DHS method and the GPU computing engine of the NEURON simulator and demonstrate applications of DeepDendrite in neuroscience tasks. We investigate how spatial patterns of spine inputs affect neuronal excitability in a detailed human pyramidal neuron model with 25,000 spines. Furthermore, we provide a brief discussion on the potential of DeepDendrite for AI, specifically highlighting its ability to enable the efficient training of biophysically detailed models in typical image classification tasks.

A Standards Organization for Open and FAIR Neuroscience: the International Neuroinformatics Coordinating Facility.

There is great need for coordination around standards and best practices in neuroscience to support efforts to make neuroscience a data-centric discipline. Major brain initiatives launched around the world are poised to generate huge stores of neuroscience data. At the same time, neuroscience, like many domains in biomedicine, is confronting the issues of transparency, rigor, and reproducibility. Widely used, validated standards and best practices are key to addressing the challenges in both big and small data science, as they are essential for integrating diverse data and for developing a robust, effective, and sustainable infrastructure to support open and reproducible neuroscience. However, developing community standards and gaining their adoption is difficult. The current landscape is characterized both by a lack of robust, validated standards and a plethora of overlapping, underdeveloped, untested and underutilized standards and best practices. The International Neuroinformatics Coordinating Facility (INCF), an independent organization dedicated to promoting data sharing through the coordination of infrastructure and standards, has recently implemented a formal procedure for evaluating and endorsing community standards and best practices in support of the FAIR principles. By formally serving as a standards organization dedicated to open and FAIR neuroscience, INCF helps evaluate, promulgate, and coordinate standards and best practices across neuroscience. Here, we provide an overview of the process and discuss how neuroscience can benefit from having a dedicated standards body.

Basal Ganglia-A Motion Perspective.

The basal ganglia represent an ancient part of the nervous system that have remained organized in a similar way over the last 500 million years and are of importance for our ability to determine which actions to choose at any given moment in time. Salient or reward stimuli act via the dopamine system and contribute to motor or procedural learning (reinforcement learning). The input stage of the basal ganglia, the striatum, is shaped by glutamatergic input from the cortex and thalamus and by the dopamine system. All intrinsic neurons of the striatum are GABAergic and inhibitory except for the cholinergic interneurons. Too little dopamine and all vertebrates show symptoms similar to that of a Parkinsonian patient, whereas too much dopamine results in hyperkinesia with involuntary movements. In this article, we discuss the detailed organization of the basal ganglia, with the different cell types, their properties, and contributions to basal ganglia functions. The striatal projection neurons represent 95% of all neurons in the striatum and are subdivided into two types, one that projects directly to the output stage, referred to as the "direct" pathway that promotes action, and the other subtype that targets the output nuclei via intercalated basal ganglia nuclei. This "indirect" pathway has an opposite effect. The striatal projection neurons express a set of ion channels that give them a high threshold for activation, whereas neurons in all other parts of the basal ganglia have a resting discharge that allows for modulation in both an increased and decreased direction. © 2020 American Physiological Society. Compr Physiol 10:1241-1275, 2020.

Cell-type-specific inhibition of the dendritic plateau potential in striatal spiny projection neurons.

Striatal spiny projection neurons (SPNs) receive convergent excitatory synaptic inputs from the cortex and thalamus. Activation of spatially clustered and temporally synchronized excitatory inputs at the distal dendrites could trigger plateau potentials in SPNs. Such supralinear synaptic integration is crucial for dendritic computation. However, how plateau potentials interact with subsequent excitatory and inhibitory synaptic inputs remains unknown. By combining computational simulation, two-photon imaging, optogenetics, and dual-color uncaging of glutamate and GABA, we demonstrate that plateau potentials can broaden the spatiotemporal window for integrating excitatory inputs and promote spiking. The temporal window of spiking can be delicately controlled by GABAergic inhibition in a cell-type-specific manner. This subtle inhibitory control of plateau potential depends on the location and kinetics of the GABAergic inputs and is achieved by the balance between relief and reestablishment of NMDA receptor Mg2+ block. These findings represent a mechanism for controlling spatiotemporal synaptic integration in SPNs.

Multirate method for co-simulation of electrical-chemical systems in multiscale modeling.

Multiscale modeling by means of co-simulation is a powerful tool to address many vital questions in neuroscience. It can for example be applied in the study of the process of learning and memory formation in the brain. At the same time the co-simulation technique makes it possible to take advantage of interoperability between existing tools and multi-physics models as well as distributed computing. However, the theoretical basis for multiscale modeling is not sufficiently understood. There is, for example, a need of efficient and accurate numerical methods for time integration. When time constants of model components are different by several orders of magnitude, individual dynamics and mathematical definitions of each component all together impose stability, accuracy and efficiency challenges for the time integrator. Following our numerical investigations in Brocke et al. (Frontiers in Computational Neuroscience, 10, 97, 2016), we present a new multirate algorithm that allows us to handle each component of a large system with a step size appropriate to its time scale. We take care of error estimates in a recursive manner allowing individual components to follow their discretization time course while keeping numerical error within acceptable bounds. The method is developed with an ultimate goal of minimizing the communication between the components. Thus it is especially suitable for co-simulations. Our preliminary results support our confidence that the multirate approach can be used in the class of problems we are interested in. We show that the dynamics ofa communication signal as well as an appropriate choice of the discretization order between system components may have a significant impact on the accuracy of the coupled simulation. Although, the ideas presented in the paper have only been tested on a single model, it is likely that they can be applied to other problems without loss of generality. We believe that this work may significantly contribute to the establishment of a firm theoretical basis and to the development of an efficient computational framework for multiscale modeling and simulations.

Segregation and crosstalk of D1 receptor-mediated activation of ERK in striatal medium spiny neurons upon acute administration of psychostimulants.

The convergence of corticostriatal glutamate and dopamine from the midbrain in the striatal medium spiny neurons (MSN) triggers synaptic plasticity that underlies reinforcement learning and pathological conditions such as psychostimulant addiction. The increase in striatal dopamine produced by the acute administration of psychostimulants has been found to activate not only effectors of the AC5/cAMP/PKA signaling cascade such as GluR1, but also effectors of the NMDAR/Ca(2+)/RAS cascade such as ERK. The dopamine-triggered effects on both these cascades are mediated by D1R coupled to Golf but while the phosphorylation of GluR1 is affected by reductions in the available amount of Golf but not of D1R, the activation of ERK follows the opposite pattern. This segregation is puzzling considering that D1R-induced Golf activation monotonically increases with DA and that there is crosstalk from the AC5/cAMP/PKA cascade to the NMDAR/Ca(2+)/RAS cascade via a STEP (a tyrosine phosphatase). In this work, we developed a signaling model which accounts for this segregation based on the assumption that a common pool of D1R and Golf is distributed in two D1R/Golf signaling compartments. This model integrates a relatively large amount of experimental data for neurons in vivo and in vitro. We used it to explore the crosstalk topologies under which the sensitivities of the AC5/cAMP/PKA signaling cascade to reductions in D1R or Golf are transferred or not to the activation of ERK. We found that the sequestration of STEP by its substrate ERK together with the insensitivity of STEP activity on targets upstream of ERK (i.e. Fyn and NR2B) to PKA phosphorylation are able to explain the experimentally observed segregation. This model provides a quantitative framework for simulation based experiments to study signaling required for long term potentiation in MSNs.

Signal enhancement in the output stage of the basal ganglia by synaptic short-term plasticity in the direct, indirect, and hyperdirect pathways.

Many of the synapses in the basal ganglia display short-term plasticity. Still, computational models have not yet been used to investigate how this affects signaling. Here we use a model of the basal ganglia network, constrained by available data, to quantitatively investigate how synaptic short-term plasticity affects the substantia nigra reticulata (SNr), the basal ganglia output nucleus. We find that SNr becomes particularly responsive to the characteristic burst-like activity seen in both direct and indirect pathway striatal medium spiny neurons (MSN). As expected by the standard model, direct pathway MSNs are responsible for decreasing the activity in SNr. In particular, our simulations indicate that bursting in only a few percent of the direct pathway MSNs is sufficient for completely inhibiting SNr neuron activity. The standard model also suggests that SNr activity in the indirect pathway is controlled by MSNs disinhibiting the subthalamic nucleus (STN) via the globus pallidus externa (GPe). Our model rather indicates that SNr activity is controlled by the direct GPe-SNr projections. This is partly because GPe strongly inhibits SNr but also due to depressing STN-SNr synapses. Furthermore, depressing GPe-SNr synapses allow the system to become sensitive to irregularly firing GPe subpopulations, as seen in dopamine depleted conditions, even when the GPe mean firing rate does not change. Similar to the direct pathway, simulations indicate that only a few percent of bursting indirect pathway MSNs can significantly increase the activity in SNr. Finally, the model predicts depressing STN-SNr synapses, since such an assumption explains experiments showing that a brief transient activation of the hyperdirect pathway generates a tri-phasic response in SNr, while a sustained STN activation has minor effects. This can be explained if STN-SNr synapses are depressing such that their effects are counteracted by the (known) depressing GPe-SNr inputs.

GABAergic circuits control spike-timing-dependent plasticity.

The spike-timing-dependent plasticity (STDP), a synaptic learning rule for encoding learning and memory, relies on relative timing of neuronal activity on either side of the synapse. GABAergic signaling has been shown to control neuronal excitability and consequently the spike timing, but whether GABAergic circuits rule the STDP remained unknown. Here we show that GABAergic signaling governs the polarity of STDP, because blockade of GABAA receptors was able to completely reverse the temporal order of plasticity at corticostriatal synapses in rats and mice. GABA controls the polarity of STDP in both striatopallidal and striatonigral output neurons. Biophysical simulations and experimental investigations suggest that GABA controls STDP polarity through depolarizing effects at distal dendrites of striatal output neurons by modifying the balance of two calcium sources, NMDARs and voltage-sensitive calcium channels. These findings establish a central role for GABAergic circuits in shaping STDP and suggest that GABA could operate as a Hebbian/anti-Hebbian switch.

The effects of NMDA subunit composition on calcium influx and spike timing-dependent plasticity in striatal medium spiny neurons.

Calcium through NMDA receptors (NMDARs) is necessary for the long-term potentiation (LTP) of synaptic strength; however, NMDARs differ in several properties that can influence the amount of calcium influx into the spine. These properties, such as sensitivity to magnesium block and conductance decay kinetics, change the receptor's response to spike timing dependent plasticity (STDP) protocols, and thereby shape synaptic integration and information processing. This study investigates the role of GluN2 subunit differences on spine calcium concentration during several STDP protocols in a model of a striatal medium spiny projection neuron (MSPN). The multi-compartment, multi-channel model exhibits firing frequency, spike width, and latency to first spike similar to current clamp data from mouse dorsal striatum MSPN. We find that NMDAR-mediated calcium is dependent on GluN2 subunit type, action potential timing, duration of somatic depolarization, and number of action potentials. Furthermore, the model demonstrates that in MSPNs, GluN2A and GluN2B control which STDP intervals allow for substantial calcium elevation in spines. The model predicts that blocking GluN2B subunits would modulate the range of intervals that cause long term potentiation. We confirmed this prediction experimentally, demonstrating that blocking GluN2B in the striatum, narrows the range of STDP intervals that cause long term potentiation. This ability of the GluN2 subunit to modulate the shape of the STDP curve could underlie the role that GluN2 subunits play in learning and development.

Striatal fast-spiking interneurons: from firing patterns to postsynaptic impact.

In the striatal microcircuit, fast-spiking (FS) interneurons have an important role in mediating inhibition onto neighboring medium spiny (MS) projection neurons. In this study, we combined computational modeling with in vitro and in vivo electrophysiological measurements to investigate FS cells in terms of their discharge properties and their synaptic efficacies onto MS neurons. In vivo firing of striatal FS interneurons is characterized by a high firing variability. It is not known, however, if this variability results from the input that FS cells receive, or if it is promoted by the stuttering spike behavior of these neurons. Both our model and measurements in vitro show that FS neurons that exhibit random stuttering discharge in response to steady depolarization do not show the typical stuttering behavior when they receive fluctuating input. Importantly, our model predicts that electrically coupled FS cells show substantial spike synchronization only when they are in the stuttering regime. Therefore, together with the lack of synchronized firing of striatal FS interneurons that has been reported in vivo, these results suggest that neighboring FS neurons are not in the stuttering regime simultaneously and that in vivo FS firing variability is more likely determined by the input fluctuations. Furthermore, the variability in FS firing is translated to variability in the postsynaptic amplitudes in MS neurons due to the strong synaptic depression of the FS-to-MS synapse. Our results support the idea that these synapses operate over a wide range from strongly depressed to almost fully recovered. The strong inhibitory effects that FS cells can impose on their postsynaptic targets, and the fact that the FS-to-MS synapse model showed substantial depression over extended periods of time might indicate the importance of cooperative effects of multiple presynaptic FS interneurons and the precise orchestration of their activity.

Postsynaptic signal transduction models for long-term potentiation and depression.

More than a hundred biochemical species, activated by neurotransmitters binding to transmembrane receptors, are important in long-term potentiation (LTP) and long-term depression (LTD). To investigate which species and interactions are critical for synaptic plasticity, many computational postsynaptic signal transduction models have been developed. The models range from simple models with a single reversible reaction to detailed models with several hundred kinetic reactions. In this study, more than a hundred models are reviewed, and their features are compared and contrasted so that similarities and differences are more readily apparent. The models are classified according to the type of synaptic plasticity that is modeled (LTP or LTD) and whether they include diffusion or electrophysiological phenomena. Other characteristics that discriminate the models include the phase of synaptic plasticity modeled (induction, expression, or maintenance) and the simulation method used (deterministic or stochastic). We find that models are becoming increasingly sophisticated, by including stochastic properties, integrating with electrophysiological properties of entire neurons, or incorporating diffusion of signaling molecules. Simpler models continue to be developed because they are computationally efficient and allow theoretical analysis. The more complex models permit investigation of mechanisms underlying specific properties and experimental verification of model predictions. Nonetheless, it is difficult to fully comprehend the evolution of these models because (1) several models are not described in detail in the publications, (2) only a few models are provided in existing model databases, and (3) comparison to previous models is lacking. We conclude that the value of these models for understanding molecular mechanisms of synaptic plasticity is increasing and will be enhanced further with more complete descriptions and sharing of the published models.

Modelling the molecular mechanisms of synaptic plasticity using systems biology approaches.

Synaptic plasticity is thought to underlie learning and memory, but the complexity of the interactions between the ion channels, enzymes and genes that are involved in synaptic plasticity impedes a deep understanding of this phenomenon. Computer modelling has been used to investigate the information processing that is performed by the signalling pathways involved in synaptic plasticity in principal neurons of the hippocampus, striatum and cerebellum. In the past few years, new software developments that combine computational neuroscience techniques with systems biology techniques have allowed large-scale, kinetic models of the molecular mechanisms underlying long-term potentiation and long-term depression. We highlight important advancements produced by these quantitative modelling efforts and introduce promising approaches that use advancements in live-cell imaging.

Run-time interoperability between neuronal network simulators based on the MUSIC framework.

MUSIC is a standard API allowing large scale neuron simulators to exchange data within a parallel computer during runtime. A pilot implementation of this API has been released as open source. We provide experiences from the implementation of MUSIC interfaces for two neuronal network simulators of different kinds, NEST and MOOSE. A multi-simulation of a cortico-striatal network model involving both simulators is performed, demonstrating how MUSIC can promote inter-operability between models written for different simulators and how these can be re-used to build a larger model system. Benchmarks show that the MUSIC pilot implementation provides efficient data transfer in a cluster computer with good scaling. We conclude that MUSIC fulfills the design goal that it should be simple to adapt existing simulators to use MUSIC. In addition, since the MUSIC API enforces independence of the applications, the multi-simulation could be built from pluggable component modules without adaptation of the components to each other in terms of simulation time-step or topology of connections between the modules.

Simple cellular and network control principles govern complex patterns of motor behavior.

The vertebrate central nervous system is organized in modules that independently execute sophisticated tasks. Such modules are flexibly controlled and operate with a considerable degree of autonomy. One example is locomotion generated by spinal central pattern generator networks (CPGs) that shape the detailed motor output. The level of activity is controlled from brainstem locomotor command centers, which in turn, are under the control of the basal ganglia. By using a biophysically detailed, full-scale computational model of the lamprey CPG (10,000 neurons) and its brainstem/forebrain control, we demonstrate general control principles that can adapt the network to different demands. Forward or backward locomotion and steering can be flexibly controlled by local synaptic effects limited to only the very rostral part of the network. Variability in response properties within each neuronal population is an essential feature and assures a constant phase delay along the cord for different locomotor speeds.

Modelling and sensitivity analysis of the reactions involving receptor, G-protein and effector in vertebrate olfactory receptor neurons.

A biochemical model of the receptor, G-protein and effector (RGE) interactions during transduction in the cilia of vertebrate olfactory receptor neurons (ORNs) was developed and calibrated to experimental recordings of cAMP levels and the receptor current (RC). The model describes the steps from odorant binding to activation of the effector enzyme which catalyzes the conversion of ATP to cAMP, and shows how odorant stimulation is amplified and delayed by the RGE transduction cascade. A time-dependent sensitivity analysis was performed on the model. The model output-the cAMP production rate-is particularly sensitive to a few, dominant parameters. During odorant stimulation it depends mainly on the initial density of G-proteins and the catalytic constant for cAMP production.

Gap junctions between striatal fast-spiking interneurons regulate spiking activity and synchronization as a function of cortical activity.

Striatal fast-spiking (FS) interneurons are interconnected by gap junctions into sparsely connected networks. As demonstrated for cortical FS interneurons, these gap junctions in the striatum may cause synchronized spiking, which would increase the influence that FS neurons have on spiking by the striatal medium spiny (MS) neurons. Dysfunction of the basal ganglia is characterized by changes in synchrony or periodicity, thus gap junctions between FS interneurons may modulate synchrony and thereby influence behavior such as reward learning and motor control. To explore the roles of gap junctions on activity and spike synchronization in a striatal FS population, we built a network model of FS interneurons. Each FS connects to 30-40% of its neighbors, as found experimentally, and each FS interneuron in the network is activated by simulated corticostriatal synaptic inputs. Our simulations show that the proportion of synchronous spikes in FS networks with gap junctions increases with increased conductance of the electrical synapse; however, the synchronization effects are moderate for experimentally estimated conductances. Instead, the main tendency is that the presence of gap junctions reduces the total number of spikes generated in response to synaptic inputs in the network. The reduction in spike firing is due to shunting through the gap junctions; which is minimized or absent when the neurons receive coincident inputs. Together these findings suggest that a population of electrically coupled FS interneurons may function collectively as input detectors that are especially sensitive to synchronized synaptic inputs received from the cortex.

A mathematical model of the mitochondrial NADH shuttles and anaplerosis in the pancreatic beta-cell.

The pancreatic beta-cells respond to an increased glycolytic flux by secreting insulin. The signal propagation goes via mitochondrial metabolism, which relays the signal to different routes. One route is an increased ATP production that, via ATP-sensitive K(+) (K(ATP)) channels, modulates the cell membrane potential to allow calcium influx, which triggers insulin secretion. There is also at least one other "amplifying" route whose nature is debated; possible candidates are cytosolic NADPH production or malonyl-CoA production. We have used mathematical modeling to analyze this relay system. The model comprises the mitochondrial NADH shuttles and the mitochondrial metabolism. We found robust signaling toward ATP, malonyl-CoA, and NADPH production. The signal toward NADPH production was particularly strong. Furthermore, the model reproduced the experimental findings that blocking the NADH shuttles attenuates the signaling to ATP production while retaining the rate of glucose oxidation (Eto K, Tsubamoto Y, Terauchi Y, Sugiyama T, Kishimoto T, Takahashi N, Yamauchi N, Kubota N, Murayama S, Aizawa T, Akanuma Y, Aizawa S, Kasai H, Yazaki Y, Kadowaki T. Science 283: 981-985, 1999) and provides an explanation for this apparent paradox. The model also predicts that the mitochondrial malate dehydrogenase reaction may proceed backward, toward malate production, if the activity of malic enzyme is sufficiently high. An increased fatty acid oxidation rate was found to attenuate the signaling strengths. This theoretical study has implications for our understanding of both the healthy and the diabetic beta-cell.

Transient calcium and dopamine increase PKA activity and DARPP-32 phosphorylation.

Reinforcement learning theorizes that strengthening of synaptic connections in medium spiny neurons of the striatum occurs when glutamatergic input (from cortex) and dopaminergic input (from substantia nigra) are received simultaneously. Subsequent to learning, medium spiny neurons with strengthened synapses are more likely to fire in response to cortical input alone. This synaptic plasticity is produced by phosphorylation of AMPA receptors, caused by phosphorylation of various signalling molecules. A key signalling molecule is the phosphoprotein DARPP-32, highly expressed in striatal medium spiny neurons. DARPP-32 is regulated by several neurotransmitters through a complex network of intracellular signalling pathways involving cAMP (increased through dopamine stimulation) and calcium (increased through glutamate stimulation). Since DARPP-32 controls several kinases and phosphatases involved in striatal synaptic plasticity, understanding the interactions between cAMP and calcium, in particular the effect of transient stimuli on DARPP-32 phosphorylation, has major implications for understanding reinforcement learning. We developed a computer model of the biochemical reaction pathways involved in the phosphorylation of DARPP-32 on Thr34 and Thr75. Ordinary differential equations describing the biochemical reactions were implemented in a single compartment model using the software XPPAUT. Reaction rate constants were obtained from the biochemical literature. The first set of simulations using sustained elevations of dopamine and calcium produced phosphorylation levels of DARPP-32 similar to that measured experimentally, thereby validating the model. The second set of simulations, using the validated model, showed that transient dopamine elevations increased the phosphorylation of Thr34 as expected, but transient calcium elevations also increased the phosphorylation of Thr34, contrary to what is believed. When transient calcium and dopamine stimuli were paired, PKA activation and Thr34 phosphorylation increased compared with dopamine alone. This result, which is robust to variation in model parameters, supports reinforcement learning theories in which activity-dependent long-term synaptic plasticity requires paired glutamate and dopamine inputs.