Large Number Of Neurons Research Articles

Comparison of IT Neural Response Statistics with Simulations.

Lehky et al. (2011) provided a statistical analysis on the responses of the recorded 674 neurons to 806 image stimuli in anterior inferotemporalm (AIT) cortex of two monkeys. In terms of kurtosis and Pareto tail index, they observed that the population sparseness of both unnormalized and normalized responses is always larger than their single-neuron selectivity, hence concluded that the critical features for individual neurons in primate AIT cortex are not very complex, but there is an indefinitely large number of them. In this work, we explore an “inverse problem” by simulation, that is, by simulating each neuron indeed only responds to a very limited number of stimuli among a very large number of neurons and stimuli, to assess whether the population sparseness is always larger than the single-neuron selectivity. Our simulation results show that the population sparseness exceeds the single-neuron selectivity in most cases even if the number of neurons and stimuli are much larger than several hundreds, which confirms the observations in Lehky et al. (2011). In addition, we found that the variances of the computed kurtosis and Pareto tail index are quite large in some cases, which reveals some limitations of these two criteria when used for neuron response evaluation.

Early postnatal exposure to isoflurane causes cognitive deficits and disrupts development of newborn hippocampal neurons via activation of the mTOR pathway.

Clinical and preclinical studies indicate that early postnatal exposure to anesthetics can lead to lasting deficits in learning and other cognitive processes. The mechanism underlying this phenomenon has not been clarified and there is no treatment currently available. Recent evidence suggests that anesthetics might cause persistent deficits in cognitive function by disrupting key events in brain development. The hippocampus, a brain region that is critical for learning and memory, contains a large number of neurons that develop in the early postnatal period, which are thus vulnerable to perturbation by anesthetic exposure. Using an in vivo mouse model we demonstrate abnormal development of dendrite arbors and dendritic spines in newly generated dentate gyrus granule cell neurons of the hippocampus after a clinically relevant isoflurane anesthesia exposure conducted at an early postnatal age. Furthermore, we find that isoflurane causes a sustained increase in activity in the mechanistic target of rapamycin pathway, and that inhibition of this pathway with rapamycin not only reverses the observed changes in neuronal development, but also substantially improves performance on behavioral tasks of spatial learning and memory that are impaired by isoflurane exposure. We conclude that isoflurane disrupts the development of hippocampal neurons generated in the early postnatal period by activating a well-defined neurodevelopmental disease pathway and that this phenotype can be reversed by pharmacologic inhibition.

Collective behavior of large-scale neural networks with GPU acceleration.

In this paper, the collective behaviors of a small-world neuronal network motivated by the anatomy of a mammalian cortex based on both Izhikevich model and Rulkov model are studied. The Izhikevich model can not only reproduce the rich behaviors of biological neurons but also has only two equations and one nonlinear term. Rulkov model is in the form of difference equations that generate a sequence of membrane potential samples in discrete moments of time to improve computational efficiency. These two models are suitable for the construction of large scale neural networks. By varying some key parameters, such as the connection probability and the number of nearest neighbor of each node, the coupled neurons will exhibit types of temporal and spatial characteristics. It is demonstrated that the implementation of GPU can achieve more and more acceleration than CPU with the increasing of neuron number and iterations. These two small-world network models and GPU acceleration give us a new opportunity to reproduce the real biological network containing a large number of neurons.

NeuroTessMesh: A Tool for the Generation and Visualization of Neuron Meshes and Adaptive On-the-Fly Refinement.

Gaining a better understanding of the human brain continues to be one of the greatest challenges for science, largely because of the overwhelming complexity of the brain and the difficulty of analyzing the features and behavior of dense neural networks. Regarding analysis, 3D visualization has proven to be a useful tool for the evaluation of complex systems. However, the large number of neurons in non-trivial circuits, together with their intricate geometry, makes the visualization of a neuronal scenario an extremely challenging computational problem. Previous work in this area dealt with the generation of 3D polygonal meshes that approximated the cells’ overall anatomy but did not attempt to deal with the extremely high storage and computational cost required to manage a complex scene. This paper presents NeuroTessMesh, a tool specifically designed to cope with many of the problems associated with the visualization of neural circuits that are comprised of large numbers of cells. In addition, this method facilitates the recovery and visualization of the 3D geometry of cells included in databases, such as NeuroMorpho, and provides the tools needed to approximate missing information such as the soma’s morphology. This method takes as its only input the available compact, yet incomplete, morphological tracings of the cells as acquired by neuroscientists. It uses a multiresolution approach that combines an initial, coarse mesh generation with subsequent on-the-fly adaptive mesh refinement stages using tessellation shaders. For the coarse mesh generation, a novel approach, based on the Finite Element Method, allows approximation of the 3D shape of the soma from its incomplete description. Subsequently, the adaptive refinement process performed in the graphic card generates meshes that provide good visual quality geometries at a reasonable computational cost, both in terms of memory and rendering time. All the described techniques have been integrated into NeuroTessMesh, available to the scientific community, to generate, visualize, and save the adaptive resolution meshes.

Distributed synaptic weights in a LIF neural network and learning rules

Leaky integrate-and-fire (LIF) models are mean-field limits, with a large number of neurons, used to describe neural networks. We consider inhomogeneous networks structured by a connectivity parameter (strengths of the synaptic weights) with the effect of processing the input current with different intensities.We first study the properties of the network activity depending on the distribution of synaptic weights and in particular its discrimination capacity. Then, we consider simple learning rules and determine the synaptic weight distribution it generates. We outline the role of noise as a selection principle and the capacity to memorize a learned signal.

Oxide-based synaptic transistors gated by solution-processed gelatin electrolytes

In human brain, a large number of neurons are connected via synapses. Simulation of the synaptic behaviors using electronic devices is the most important step for neuromorphic systems. In this paper, proton conducting gelatin electrolyte-gated oxide field-effect transistors (FETs) were used for emulating synaptic functions, in which the gate electrode is regarded as pre-synaptic neuron and the channel layer as the post-synaptic neuron. In analogy to the biological synapse, a potential spike can be applied at the gate electrode and trigger ionic motion in the gelatin electrolyte, which in turn generates excitatory post-synaptic current (EPSC) in the channel layer. Basic synaptic behaviors including spike time-dependent EPSC, paired-pulse facilitation (PPF), self-adaptation, and frequency-dependent synaptic transmission were successfully mimicked. Such ionic/electronic hybrid devices are beneficial for synaptic electronics and brain-inspired neuromorphic systems.

Superconducting Optoelectronic Circuits for Neuromorphic Computing

Neural networks have proven effective for solving many difficult computational problems. Implementing complex neural networks in software is very computationally expensive. To explore the limits of information processing, it will be necessary to implement new hardware platforms with large numbers of neurons, each with a large number of connections to other neurons. Here we propose a hybrid semiconductor-superconductor hardware platform for the implementation of neural networks and large-scale neuromorphic computing. The platform combines semiconducting few-photon light-emitting diodes with superconducting-nanowire single-photon detectors to behave as spiking neurons. These processing units are connected via a network of optical waveguides, and variable weights of connection can be implemented using several approaches. The use of light as a signaling mechanism overcomes fanout and parasitic constraints on electrical signals while simultaneously introducing physical degrees of freedom which can be employed for computation. The use of supercurrents achieves the low power density necessary to scale to systems with enormous entropy. The proposed processing units can operate at speeds of at least $20$ MHz with fully asynchronous activity, light-speed-limited latency, and power densities on the order of 1 mW/cm$^2$ for neurons with 700 connections operating at full speed at 2 K. The processing units achieve an energy efficiency of $\approx 20$ aJ per synapse event. By leveraging multilayer photonics with deposited waveguides and superconductors with feature sizes $>$ 100 nm, this approach could scale to systems with massive interconnectivity and complexity for advanced computing as well as explorations of information processing capacity in systems with an enormous number of information-bearing microstates.

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations.

This protocol outlines large-scale reconstructions of neurons combined with the use of independent and unbiased clustering analyses to create a comprehensive survey of the morphological characteristics observed among a selective neuronal population. Combination of these techniques constitutes a novel approach for the collection and analysis of neuroanatomical data. Together, these techniques enable large-scale, and therefore more comprehensive, sampling of selective neuronal populations and establish unbiased quantitative methods for describing morphologically unique neuronal classes within a population. The protocol outlines the use of modified rabies virus to selectively label neurons. G-deleted rabies virus acts like a retrograde tracer following stereotaxic injection into a target brain structure of interest and serves as a vehicle for the delivery and expression of EGFP in neurons. Large numbers of neurons are infected using this technique and express GFP throughout their dendrites, producing "Golgi-like" complete fills of individual neurons. Accordingly, the virus-mediated retrograde tracing method improves upon traditional dye-based retrograde tracing techniques by producing complete intracellular fills. Individual well-isolated neurons spanning all regions of the brain area under study are selected for reconstruction in order to obtain a representative sample of neurons. The protocol outlines procedures to reconstruct cell bodies and complete dendritic arborization patterns of labeled neurons spanning multiple tissue sections. Morphological data, including positions of each neuron within the brain structure, are extracted for further analysis. Standard programming functions were utilized to perform independent cluster analyses and cluster evaluations based on morphological metrics. To verify the utility of these analyses, statistical evaluation of a cluster analysis performed on 160 neurons reconstructed in the thalamic reticular nucleus of the thalamus (TRN) of the macaque monkey was made. Both the original cluster analysis and the statistical evaluations performed here indicate that TRN neurons are separated into three subpopulations, each with unique morphological characteristics.

Hopf bifurcation in a nonlocal nonlinear transport equation stemming from stochastic neural dynamics.

In this work, we provide three different numerical evidences for the occurrence of a Hopf bifurcation in a recently derived [De Masi et al., J. Stat. Phys. 158, 866-902 (2015) and Fournier and löcherbach, Ann. Inst. H. Poincaré Probab. Stat. 52, 1844-1876 (2016)] mean field limit of a stochastic network of excitatory spiking neurons. The mean field limit is a challenging nonlocal nonlinear transport equation with boundary conditions. The first evidence relies on the computation of the spectrum of the linearized equation. The second stems from the simulation of the full mean field. Finally, the last evidence comes from the simulation of the network for a large number of neurons. We provide a "recipe" to find such bifurcation which nicely complements the works in De Masi et al. [J. Stat. Phys. 158, 866-902 (2015)] and Fournier and löcherbach [Ann. Inst. H. Poincaré Probab. Stat. 52, 1844-1876 (2016)]. This suggests in return to revisit theoretically these mean field equations from a dynamical point of view. Finally, this work shows how the noise level impacts the transition from asynchronous activity to partial synchronization in excitatory globally pulse-coupled networks.

Quantification of cortical neurons: a neuroanatomical myth in neuroscience

AbstractThe objective was to analyze whether there was a definite number of cerebral cortical neurons for humans. Methodology: data was examined for the total number of brain neurons in the literature related to the field of neuroscience. This suggested that the human cerebral cortex has a large number of neurons whose value changes with the life cycle due to several factors that lead to have an indefinite number. Discussion: an analysis based on the literature showed the main factors influencing the total number of cortical neurons such as neurogenesis, apoptosis during life cycle, gender and pathological conditions. The inconsistent assessment of the number of neurons in the human cortex has been discussed by several authors, emphasizing that the values that have been displayed in the literature are presumably uncertain and are based on the views of some authors or the conceptions of others without carrying out replicable scientific studies. The above described has exhibited variable and inconsistent values in different texts about this issue. It has been concluded that the total number of cortical neurons varies due to diverse factors that play a role in determining the number of neurons in the human cerebral cortex.

Morphological relationship between the superior cervical ganglion and cervical nerves in Japanese cadaver donors.

IntroductionThere are various communications between the superior cervical ganglion (SCG) and the vagus and glossopharyngeal nerves. However, little information exists concerning the origin of these sympathetic ganglion branches at the superior, middle, and inferior regions of the human SCG. The aim of this study was to describe the human SCG in a morphometric manner with the communication with cranial and cervical nerves and supply.MethodsThis study characterized 72 SCG samples from 54 elderly Japanese human cadavers (30 males, 24 females; 65–100 years old). The SCG size (length, width, and thickness) and location were measured from the jugular foramen. We also defined the communication branches of the SCG to the vagus, glossopharyngeal, cervical, and accessory nerves at three regions (superior, middle, and inferior regions) of the SCG. Finally, we examined the arrangement and origin of the branch communications in detail and confirmed our observations, using histological sections of the SCG.ResultsThe SCG in all cadaver donors was detected at the C2 and C3 vertebra levels. The number of SCG branches supplied the communicating branches, such as the carotid branch, communicating branch of the vagus nerve, and glossopharyngeal nerve, were frequently detected in the superior region of the SCG (χ2 = 587.72, df = 26, p < .001). The number of ganglion cells with a large number of neurons per unit area (1 mm2) was most often found in the middle region with shrunken neurons of the SCG compared with other regions.ConclusionThe communication branches of the SCG are mainly connected to the vagus and glossopharyngeal nerves. Characterizing these branches can provide useful data for head and neck ganglion block and surgical treatments.

Machine-learned cluster identification in high-dimensional data

BackgroundHigh-dimensional biomedical data are frequently clustered to identify subgroup structures pointing at distinct disease subtypes. It is crucial that the used cluster algorithm works correctly. However, by imposing a predefined shape on the clusters, classical algorithms occasionally suggest a cluster structure in homogenously distributed data or assign data points to incorrect clusters. We analyzed whether this can be avoided by using emergent self-organizing feature maps (ESOM). MethodsData sets with different degrees of complexity were submitted to ESOM analysis with large numbers of neurons, using an interactive R-based bioinformatics tool. On top of the trained ESOM the distance structure in the high dimensional feature space was visualized in the form of a so-called U-matrix. Clustering results were compared with those provided by classical common cluster algorithms including single linkage, Ward and k-means. ResultsWard clustering imposed cluster structures on cluster-less “golf ball”, “cuboid” and “S-shaped” data sets that contained no structure at all (random data). Ward clustering also imposed structures on permuted real world data sets. By contrast, the ESOM/U-matrix approach correctly found that these data contain no cluster structure. However, ESOM/U-matrix was correct in identifying clusters in biomedical data truly containing subgroups. It was always correct in cluster structure identification in further canonical artificial data. Using intentionally simple data sets, it is shown that popular clustering algorithms typically used for biomedical data sets may fail to cluster data correctly, suggesting that they are also likely to perform erroneously on high dimensional biomedical data. ConclusionsThe present analyses emphasized that generally established classical hierarchical clustering algorithms carry a considerable tendency to produce erroneous results. By contrast, unsupervised machine-learned analysis of cluster structures, applied using the ESOM/U-matrix method, is a viable, unbiased method to identify true clusters in the high-dimensional space of complex data.

Scaling Properties of Dimensionality Reduction for Neural Populations and Network Models.

Recent studies have applied dimensionality reduction methods to understand how the multi-dimensional structure of neural population activity gives rise to brain function. It is unclear, however, how the results obtained from dimensionality reduction generalize to recordings with larger numbers of neurons and trials or how these results relate to the underlying network structure. We address these questions by applying factor analysis to recordings in the visual cortex of non-human primates and to spiking network models that self-generate irregular activity through a balance of excitation and inhibition. We compared the scaling trends of two key outputs of dimensionality reduction—shared dimensionality and percent shared variance—with neuron and trial count. We found that the scaling properties of networks with non-clustered and clustered connectivity differed, and that the in vivo recordings were more consistent with the clustered network. Furthermore, recordings from tens of neurons were sufficient to identify the dominant modes of shared variability that generalize to larger portions of the network. These findings can help guide the interpretation of dimensionality reduction outputs in regimes of limited neuron and trial sampling and help relate these outputs to the underlying network structure.

Spatiotemporal Behavior of Small-World Neuronal Networks Using a Map-Based Model

In this paper, the dynamics of a map-based model is proposed firstly. It is a simple model which can not only reproduce rich behaviors of biological neurons but also compute efficiently. Then, the dynamics of two coupled maps that model the behavior of two electrically coupled neurons is discussed. By tuning the coupling strength, synchronization of two spiking or bursting neurons are simulated. Furthermore, the spatiotemporal behavior of a small-world neuronal network using map-based model is studied. Detailed investigations reveal that the collective dynamics of neuronal activity will be affected by varying some key parameters, such as the coupling strength of neurons, connection probability and the number of nearest neighbor in small-world topology. Our study suggests that the map-based model will give us a new opportunity to reproduce the real biological network containing a large number of neurons.

Вычисление областей устойчивости дискретных моделей больших нейронных сетей типа small world

The article is devoted to description of discrete models of small world networks with a large number of neurons with a certain parameter p varying from 0 to 1. For p=0 have model, regular neural networks, which is a ring network in which each neuron interacts with several neighbors on the ring. In the case p=1 have a model with randomly distributed connections. When the values of $p$ not exceeding 0.1 have the Watts-Strogatz small world network. Such a neural network can be models of different neural structures in living organisms, for example, the hippocampus of the mammalian brain. This paper examines the dynamics of change areas of stability of such neural networks when 0\leq p \leq 0.1. Numerical experiments show an increase in sustainability in the transition from a regular network to small world.

Expression of the Serotonin Transporter in the Ventrolateral Part of the Solitary Tract Nucleus in Rats during the Early Postnatal Period in Normal Conditions and in Serotoninergic System Deficiency during the Prenatal Period of Development

An immunocytochemical study method was used to investigate the expression of the serotonin transporter (5-HTT) in the ventrolateral part of the solitary tract nucleus in Wistar rats during the early postnatal period (days 5 and 10) in normal conditions (n = 10) and in conditions of prenatal serotonin deficiency (n = 12). During the early postnatal period, the ventrolateral section of the solitary tract nucleus was found to show transient expression of 5-HTT, most marked in the caudal part of the ventral subnucleus. The numbers of neurons synthesizing 5-HTT in the rostral part of the ventral and lateral subnuclei were small and did not change with age. The caudal part of the ventral subnucleus contained a large number of neurons synthesizing 5-HTT on day 5 of the postnatal period, and this number decreased significantly with age. The caudal part of the lateral subnucleus contained a small number of cells expressing 5-HTT, and this number also decreased with age. 5-HTT expression levels were significantly higher in the caudal areas of both the ventral and lateral subnuclei than in the rostral areas. Prenatal serotonin deficiency decreased the numbers of neurons synthesizing 5-HTT in the nuclei studied here.

Ginsenosides Rg1 enhance the ability of olfactory ensheathing cells to repair spinal cord injury: An experimental study

Objective To study the effects of ginsenoside Rg1 on effects of olfactory ensheathing cells (OECs) in spinal cord injury as well as their expression of neurotrophic factors like ciliary neurotrophic factor (CNTF), vascular endothelial growth factor (VEGF). Methods OECs were primary cultured and purified from olfactory bulb of newborn SD rats. Purified cells were divided into 2 groups, the experimental group was treated with Rg1 in concentration of 40 μg/ml for 72 h (the optimal concentration and action time which have been proven by our previous study), the control group received an equal volume of Dulbecco's Modified Eagle Media: Nutrient Mixture F-12 (DMEM/F12) medium. The models of spinal cord injury was established by using Allen's combat, and were randomly divided into 3 groups, OECs group (inject the control group cells), Rg1+ OECs group (inject the experimental group cells) and control group (inject an equal volume of DMEM/F12 medium). Using reverse transcriptase-polymerase chain reaction (RT-PCR) to determine mRNA expression of CNTF and VEGF in OECs group and Rg1+ OECs group. Results The Basso Beattie Bresnahan (BBB) score of Rg1+ OECs group was higher than the other two groups (15.0±1.0 vs.12.0±2.0, 7.4±0.9, P<0.05). Histology revealed that Rg1+ OECs group had larger number of neurons (64±3, 38±2, 26±2, P<0.01), lower level of GFAP expression and less damage of spinal cord tissue than OECs group and control group (8.65±0.32, 4.93±0.25, 2.21±0.17, P<0.01). RT-PCR analysis demonstrated that Rg1 significantly up-regulated mRNA expression of CNTF and VEGF (0.78±0.03, 0.39±0.03; 0.86±0.03, 0.61±0.02, P<0.05). Conclusion Ginsinoside Rg1 obviously improve the effects of olfactory ensheathing cell transplantation for spinal cord injury by the up-regulation of CNTF and VEGF. Key words: Ginsenoside Rg1; Olfactory ensheathing cells; Spinal cord injury; Cell transplantation; Gene expression

Editorial: Mechanisms of Neuronal Migration during Corticogenesis.

Editorial: Mechanisms of Neuronal Migration during Corticogenesis.

Immunolocalization of gaseous neurotransmitters in interneurons in the medulla oblongata in rats

We studied groups of internuclear interneurons (IN) that are located between the magnocellular and parvocellular reticular nuclei (IN1), reticular parvocellular nucleus, the nucleus of the solitary tract (IN2), and around the reticular lateral nucleus (IN3) that express the neuronal NO-synthase, cystathionine s-synthase, or heme oxygenase-2 in the medulla oblongata of Wistar rats using histochemical and immunohistochemical methods. We found local differences in the organization of INs. These differences were evident between IN1 cells, a cell group that consists of a relatively large number of large neurons, including giant cells, and IN2–IN3 cells. In interneurons of the latter two IN groups, we found significantly higher concentrations of cells and intensity of immunohistochemical staining. The local differences between the cells that contain cystathionine s-synthase were less pronounced than between other gaseous neurotransmitter interneurons. These data indicate that the internuclear interneurons have the structure that is necessary for the successful management of integration processes in the brain.

Spike sorting for large, dense electrode arrays.

Developments in microfabrication technology have enabled the production of neural electrode arrays with hundreds of closely-spaced recording sites, and electrodes with thousands of sites are currently under development. These probes in principle allow the simultaneous recording of very large numbers of neurons. However, use of this technology requires the development of techniques for decoding the spike times of the recorded neurons, from the raw data captured from the probes. Here, we present a set of novel tools to solve this problem, implemented in a suite of practical, user-friendly, open-source software. We validate these methods on data from the cortex, hippocampus, and thalamus of rat, mouse, macaque, and marmoset, demonstrating error rates as low as 5%.