Sequential Spikes Research Articles

Modeling and Analysis of Spike Signal Sequence for Memristor Crossbar Array in Neuromorphic Chips

This paper presents the efficient systematic methods for modeling and analysis of spike signal sequence in crossbar arrays for neuromorphic computing chips. A novel spike signal sequence is proposed, where the ideal spike sequence with only spike time information in the original spiking neural network (SNN) algorithm is mapped onto actual spike waveform by stitching neighboring sequential spikes together with certain overlaps. We thoroughly investigate and analyze the performance of the input encoding as well as the implementation of spike timing dependent plasticity (STDP)-based SNN on memristor crossbar arrays with the proposed spike signal sequence. A detailed circuit model of a crossbar array, consisting of resistance, capacitance and inductance derived by the partial equivalent element circuit (PEEC) method, is created to simulate the training process of SNN. The proposed spike signal sequence is demonstrated that is able to achieve accurate input encoding as well as high recognition accuracy when it is used to perform the classification task on MNIST handwritten digits. The spike signal sequence is further analyzed and assessed in terms of the main factors affecting its encoding accuracy and the parasitic effects of crossbar arrays on its robustness.

Onion-Like nanoparticles of the metal–organic framework UiO-66 synthesized by sequential spike crystal growth

Reproducible size and morphology of metal–organic framework (MOF) particles are essential for tuning these materials for applications like heterogeneous catalysis. Particle size measurements from electron micrographs provide information about the impact of changing synthetic parameters on MOF growth, but reported data are usually limited to averages and statistical distributions. Further elucidating synthetic control parameters could facilitate studies of facet-dependent and non-classical growth mechanisms. In this work, phase-pure nanoparticles of the MOF UiO-66 were synthesized using alternating spikes of zirconium and hafnium-based nodes, forming an onion-like structure with alternating layers of Zr-UiO-66 and Hf-UiO-66. In conventional, bright-field transmission electron micrographs, Hf-UiO-66 and Zr-UiO-66 layers appeared darker and lighter, respectively. Alternating layers of Hf-UiO-66 and Zr-UiO-66 were also apparent in elemental maps obtained using scanning transmission electron microscopy with energy-dispersive X-ray spectroscopy. Average particle sizes increased monotonically with each spike, and no Zr-only or Hf-only particles were observed after the first spike, indicating that new material was consistently incorporated onto existing seeds. Surface area of the onion-like particles and their uptake of Ni2+ from solution were both between the values measured for Zr-UiO-66 and Hf-UiO-66. Results demonstrate the use of sequential spikes of nodes in monitoring crystal growth in MOFs.

P.097 After-discharges and presurgical cortical stimulation in stereo-encephalography in the study of drug-resistant epilepsy

Background: Background: Cortical stimulation (CS) as a part of presurgical investigations in patients undergoing implantation of depth electrodes (SEEG) is a growing practice in some Comprehensive Epilepsy Centers. After-discharges (AD) are useful to determine epileptogenic tissue within or outside the epileptogenic network. Classification of afterdischarges was proposed by Blume using subdural recordings(1); its utility in SEEG is unknown. Methods: Methods: Single center, retrospective study that included patients with SEEG that underwent CS in the Epilepsy Monitoring Unit. Demographic characteristics were explored and Blume’s proposed AD classification was used to determine whether or not the CS changed surgical outcomes. Results: Results: From January 2015 to June 2021, a total of 177 patients were implanted with SEEG and analyzed. 95 patients had CS and 91 had AD. Morphologies found were: Rhythmic waves in 4 (0.04%), Rhythmic waves evolving into spikes in18 (19%), Polyspike bursts in 14 (15%), Spike-waves in 28 (30%), and sequential spikes in 18 (19%). 12/14 (86%) patients with Spike-waves had Engel I outcome; Engel class IV patients were more likely to have an evolution of morphology and frequency of ADs in 3 patients(75%). Conclusions: Conclusions: The most frequent morphology of ADs seen was spike-waves. AD morphology and duration may predict post-operative seizure outcomes.

Afterdischarges elicited by cortical electric stimulation in humans: When do they occur and what do they mean?

IntroductionAfterdischarges (ADs) are a common and unwanted byproduct of direct cortical stimulation during invasive electroencephalography (EEG) recordings. Brief pulse stimulation (BPS) can sometimes terminate ADs. This study investigated AD characteristics and their relevance for emergence of stimulation seizures. In addition, AD response to BPS was analyzed. Material and methodsInvasive EEG recordings including mapping with direct cortical stimulation in patients with refractory epilepsy at the Erlangen Epilepsy Center were retrospectively reviewed. Afterdischarge defined as stimulation-induced rhythmic epileptiform discharges of more than a two-second duration were analyzed regarding incidence, localization, duration, propagation pattern, morphology, and seizure emergence. In addition, the influence of AD characteristics and stimulation settings on BPS success rate was studied. ResultsA number of 4261 stimulation trials in 20 patients were investigated. Afterdischarge occurred in 518 trials (14.2%) and lasted 12.4 s (standard deviation [SD]: 8.6 s) on average. We elicited ADs in the seizure onset zone (SOZ) (n = 64; 19.4%), the irritative zone (n = 105, 20.0%), and outside the irritative area (n = 222, 12.5%). Rhythmic spikes (30.5%) and spike–wave complexes (30.3%) represented predominant morphologies. Afterdischarge morphology in the SOZ and hippocampus differed from other areas with polyspikes and sequential spikes being the most common types there (p = 0.0005; p < 0.0001 respectively). Hippocampal ADs were particularly frequent (n = 50, 38.2%) and long-lasting (mean: 16.6, SD: 8.3 s). Brief pulse stimulation was applied in 18.1% of the AD trials (n = 94) and was successful in 37.4% (n = 40). Success rates were highest when BPS was delivered within 9.5 s (p = 0.0048) and in ADs of spike–wave morphology (p = 0.0004). Fifteen clinical seizures emerged from ADs (3.55%), mostly evolving from sequential spikes. Afterdischarges in patients with stimulation seizures appeared more widespread (p < 0.0001) and lasted longer (mean duration 7.0 s) than in those without (mean duration 21.0 s, p = 0.0054). ConclusionAfterdischarges appear in the epileptogenic and nonepileptogenic cortex. Duration and propagation patterns can help to quantify the risk of stimulation seizures, with sequential spikes being most susceptible to seizure elucidation. The hippocampus is highly sensitive to AD release. Brief pulse stimulation is a safe and efficacious way to terminate ADs, especially when delivered quickly after AD onset.

Properties of afterdischarges from electrical stimulation in patients with epilepsy

ObjectiveTo investigate the properties of afterdischarges (ADs) from intracerebral electrical stimulation (ES) in patients with epilepsy who underwent stereotactic electroencephalography (SEEG) and determine the relationship between epileptogenic zone (EZ) or irritative zone (IZ) and ADs. MethodsWe retrospectively analyzed 10 patients with intractable epilepsy who underwent SEEG. ESs were delivered following the given parameters: bipolar, biphasic, 50Hz, 0.2ms pulse duration, 0.5–10mA. The properties of ADs were documented, including their incidence, location, threshold, morphology and evolution. ResultsA total of 213 ADs (5%) were elicited by 4701 trains of ES. Stimulation through contacts implanted in the hippocampus (59%) generally evoked more ADs than contacts elsewhere (19%). AD thresholds for hippocampal stimulation were significantly lower than those for stimulation in grey matter. Polyspikes (58%) were the most common AD morphology. Evolution occurred more commonly with sequential spikes (47%) than with other AD morphologies (14%). There was no significant correlation between the location of ADs and EZ. However, ADs were significantly more frequently localized to IZ than areas outside IZ (P<0.05). ConclusionsThere seemed to be a lack of correlation between the location of ADs and EZ. However, ADs were more likely to be elicited in IZ.

Transient reduction in theta power caused by interictal spikes in human temporal lobe epilepsy.

The inhibitory impacts of spikes on LFP theta rhythms(4-8Hz) are investigated around sporadic spikes(SSs) based on intracerebral EEG of 4 REM sleep patients with temporal lobe epilepsy(TLE) under the pre-surgical monitoring. Sequential interictal spikes in both genesis area and extended propagation pathway are collected, that, SSs genesis only in anterior hippocampus(aH)(possible propagation pathway in Entorhinal cortex(EC)), only in EC(possible propagation pathway in aH), and in both aH and EC synchronously. Instantaneous theta power was estimated by using Gabor wavelet transform, and theta power level was estimated by averaged over time and frequency before SSs(350ms pre-spike) and after SSs(350ms post-spike). The inhibitory effect around spikes was evaluated by the ratio of theta power level difference between pre-spike and post-spike to pre-spike theta power level. The findings were that theta power level was reduced across SSs, and the effects were more sever in the case of SSs in both aH and EC synchronously than either SSs only in EC or SSs only in aH. It is concluded that interictal spikes impair LFP theta rhythms transiently and directly. The work suggests that the reduction of theta power after the interictal spike might be an evaluation indicator of damage of epilepsy to human cognitive rhythms.

Ischemia deteriorates spike encoding at cortical GABAergic neurons and cerebellar Purkinje cells by increasing the intracellular Ca2.

GABAergic neurons play a critical role in the central nervous system, such as well-organized behaviors. The ischemic cell death is presumably initiated by neuronal excitotoxicity resulted from the dysfunction of GABAergic neurons. It is not clear how ischemia influences different types of GABAergic neurons and whether intracellular Ca2+ plays a key role in the ischemic excitotoxicity. We have investigated this issue at cortical GABAergic neurons and cerebellar Purkinje cells by whole-cell recording in mouse brain slices, and the roles of intracellular Ca2+ are examined by BABTA infusion. Compare with the data from a group of control, ischemia causes by lowering purfusion rate lowers spike encoding at cortical GABAergic neurons and enhances encoding ability at cerebellar Purkinje cells. These differential effects of ischemia on spike encoding are mechanistically associated with the changes in the refractory periods and threshold potentials of sequential spikes. These ischemia-induced dysfunction of spike encoding at two types of GABAergic cells are prevented by BABTA infusion. Therefore, the ischemia destabilizes the spike encoding of GABAergic cells via raising intracellular Ca2+. Our findings indicate that ischemia preferentially causes the dysfunction of spike encoding at GABAergic neurons by the up-regulation of intracellular Ca2+ level, which leads to neuronal excitotoxicity.

Emergence of functional subnetworks in layer 2/3 cortex induced by sequential spikes in vivo

During cortical circuit development in the mammalian brain, groups of excitatory neurons that receive similar sensory information form microcircuits. However, cellular mechanisms underlying cortical microcircuit development remain poorly understood. Here we implemented combined two-photon imaging and photolysis in vivo to monitor and manipulate neuronal activities to study the processes underlying activity-dependent circuit changes. We found that repeated triggering of spike trains in a randomly chosen group of layer 2/3 pyramidal neurons in the somatosensory cortex triggered long-term plasticity of circuits (LTPc), resulting in the increased probability that the selected neurons would fire when action potentials of individual neurons in the group were evoked. Significant firing pattern changes were observed more frequently in the selected group of neurons than in neighboring control neurons, and the induction was dependent on the time interval between spikes, N-methyl-D-aspartate (NMDA) receptor activation, and Calcium/calmodulin-dependent protein kinase II (CaMKII) activation. In addition, LTPc was associated with an increase of activity from a portion of neighboring neurons with different probabilities. Thus, our results demonstrate that the formation of functional microcircuits requires broad network changes and that its directionality is nonrandom, which may be a general feature of cortical circuit assembly in the mammalian cortex.

Input-dependent subcellular localization of spike initiation between soma and axon at cortical pyramidal neurons

BackgroundAction potentials can be initiated at various subcellular compartments, such as axonal hillock, soma and dendrite. Mechanisms and physiological impacts for this relocation remain elusive, which may rely on input signal patterns and intrinsic properties in these subcellular compartments. We examined this hypothesis at the soma and axon of cortical pyramidal neurons by analyzing their spike capability and voltage-gated sodium channel dynamics in response to different input signals.ResultsElectrophysiological recordings were simultaneously conducted at the somata and axons of identical pyramidal neurons in the cortical slices. The somata dominantly produced sequential spikes in response to long-time steady depolarization pulse, and the axons produced more spikes in response to fluctuated pulse. Compared with the axons, the somata possessed lower spike threshold and shorter refractory periods in response to long-time steady depolarization, and somatic voltage-gated sodium channels demonstrated less inactivation and easier reactivation in response to steady depolarization. Based on local VGSC dynamics, computational simulated spike initiation locations were consistent with those from the experiments. In terms of physiological impact, this input-dependent plasticity of spike initiation location made neuronal encoding to be efficient.ConclusionsLong-time steady depolarization primarily induces somatic spikes and short-time pulses induce axonal spikes. The input signal patterns influence spike initiations at the axon or soma of cortical pyramidal neurons through modulating local voltage-gated sodium channel dynamics.

Frequency-Dependent Reliability of Spike Propagation Is Function of Axonal Voltage-Gated Sodium Channels in Cerebellar Purkinje Cells

The spike propagation on nerve axons, like synaptic transmission, is essential to ensure neuronal communication. The secure propagation of sequential spikes toward axonal terminals has been challenged in the neurons with a high firing rate, such as cerebellar Purkinje cells. The shortfall of spike propagation makes some digital spikes disappearing at axonal terminals, such that the elucidation of the mechanisms underlying spike propagation reliability is crucial to find the strategy of preventing loss of neuronal codes. As the spike propagation failure is influenced by the membrane potentials, this process is likely caused by altering the functional status of voltage-gated sodium channels (VGSC). We examined this hypothesis in Purkinje cells by using pair-recordings at their somata and axonal blebs in cerebellar slices. The reliability of spike propagation was deteriorated by elevating spike frequency. The frequency-dependent reliability of spike propagation was attenuated by inactivating VGSCs and improved by removing their inactivation. Thus, the functional status of axonal VGSCs influences the reliability of spike propagation.

Upregulation of transmitter release probability improves a conversion of synaptic analogue signals into neuronal digital spikes

Action potentials at the neurons and graded signals at the synapses are primary codes in the brain. In terms of their functional interaction, the studies were focused on the influence of presynaptic spike patterns on synaptic activities. How the synapse dynamics quantitatively regulates the encoding of postsynaptic digital spikes remains unclear. We investigated this question at unitary glutamatergic synapses on cortical GABAergic neurons, especially the quantitative influences of release probability on synapse dynamics and neuronal encoding. Glutamate release probability and synaptic strength are proportionally upregulated by presynaptic sequential spikes. The upregulation of release probability and the efficiency of probability-driven synaptic facilitation are strengthened by elevating presynaptic spike frequency and Ca2+. The upregulation of release probability improves spike capacity and timing precision at postsynaptic neuron. These results suggest that the upregulation of presynaptic glutamate release facilitates a conversion of synaptic analogue signals into digital spikes in postsynaptic neurons, i.e., a functional compatibility between presynaptic and postsynaptic partners.

A growth factor signaling cascade confined to circular ruffles in macrophages

SummaryThe formation of macropinosomes requires large-scale movements of membranes and the actin cytoskeleton. Over several minutes, actin-rich surface ruffles transform into 1–5 µm diameter circular ruffles, which close at their distal margins, creating endocytic vesicles. Previous studies using fluorescent reporters of phosphoinositides and Rho-family GTPases showed that signals generated by macrophages in response to the growth factor Macrophage Colony-Stimulating Factor (M-CSF) appeared transiently in domains of plasma membrane circumscribed by circular ruffles. To address the question of how signaling molecules are coordinated in such large domains of plasma membrane, this study analyzed the relative timing of growth factor-dependent signals as ruffles transformed into macropinosomes. Fluorescent protein chimeras expressed in macrophages were imaged by microscopy and quantified relative to circular ruffle formation and cup closure. The large size of macropinocytic cups allowed temporal resolution of the transitions in phosphoinositides and associated enzyme activities that organize cup closure. Circular ruffles contained transient and sequential spikes of phosphatidylinositol (4,5)-bisphosphate (PI(4,5)P2), phosphatidylinositol (3,4,5)-trisphosphate (PIP3), diacylglycerol, PI(3,4)P2, PI(3)P and the activities of protein kinase C-α, Rac1, Ras and Rab5. The confinement of this signal cascade to circular ruffles indicated that diffusion barriers present in these transient structures focus feedback activation and deactivation of essential enzyme activities into restricted domains of plasma membrane.

Input Signal Patterns Dominate a Plasticity of Spike-Onset Location at Cortical Pyramid Neurons through Local VGSCs

Synaptic signals drive the neurons to fire sequential spikes as digital codes. Short-term pulses initiate a spike at the axonal hillock, and physiological signals may initiate digital spikes at the soma. The regulation, mechanism and impact for spike-onset relocation between subcellular compartments remain unknown, which we investigated by simultaneously recording the soma and axon of pyramidal neurons in cortical slices. By analyzing the abilities of firing spikes, the time phases of spike-onset and the relocations of spike-initiation at these two compartments, we have found that long-time steady depolarization induces sequential spikes at the soma, but fluctuated one induces spikes at the axon. The soma in response to long-time pulses shows low thresholds and short refractory periods, or vice versa. Compared with axonal voltage-gated sodium channels (VGSC), somatic VGSCs in response to a pre-depolarization appear less inactivated and easily reactivated. Based on these VGSC features, the location of spike-onset simulated in NEURON model is consistent with the experiments. The patterns of input signals dominate spike initiations at the axon or soma of cortical pyramidal neurons through influencing local VGSC function. The plasticity of spike-onset location allows the neurons to program the brain codes economically. [This study is supported by the National Award for Outstanding Young Scientist (30325021), National Basic Research Program (2011CB504405) and Natural Science Foundation China (30870517, 30990261 and 81171033) to JHW].

Acidosis leads to brain dysfunctions through impairing cortical GABAergic neurons

Acidosis, associated with metabolic disorders, leads to the pathological changes of cognition and behavior in the clinical practices of neurology and psychology. The cellular mechanisms underlying these cerebral dysfunctions remain unclear. By using electrophysiological approach and changing extracellular pH, we have investigated the effects of acidic environment on cortical GABAergic neurons in terms of their abilities of firing spikes and responses to synaptic inputs. Artificial cerebral spinal fluid in low pH impairs the responses to excitatory synaptic inputs and the abilities of encoding sequential spikes at these GABAergic neurons. The impairments of neuronal spiking are associated with the increases of refractory periods and threshold potentials. Our studies reveal that acidosis may impair cortical GABAergic neurons and in turn deteriorate brain functions, in which their final targets are voltage-gated sodium channels and glutamate receptor-channels.

Physiological synaptic signals initiate sequential spikes at soma of cortical pyramidal neurons

The neurons in the brain produce sequential spikes as the digital codes whose various patterns manage well-organized cognitions and behaviors. A source for the physiologically integrated synaptic signals to initiate digital spikes remains unknown, which we studied at pyramidal neurons of cortical slices. In dual recordings from the soma vs. axon, the signals recorded in vivo induce somatic spikes with higher capacity, which is associated with lower somatic thresholds and shorter refractory periods mediated by voltage-gated sodium channels. The introduction of these parameters from the soma and axon into NEURON model simulates sequential spikes being somatic in origin. Physiological signals integrated from synaptic inputs primarily trigger the soma to encode neuronal digital spikes.

Acidosis and alkalosis impair brain functions through weakening spike encoding at cortical GABAergic neurons

Acidosis and alkalosis, associated with metabolic disorders, lead to the pathological changes of cognition and behaviors in clinical practices of neurology and psychology. Cellular mechanisms for these functional disorders in the central nervous system remain unclear. We have investigated the influences of acidosis and alkalosis on the functions of cortical GABAergic neurons. Both acidosis and alkalosis impair the ability of encoding sequential spikes at these GABAergic neurons. The impairments of their spiking are associated with the increases of refractory periods, threshold potential and afterhyperpolarization. Our studies reveal that acidosis and alkalosis impair cortical GABAergic neurons and in turn deteriorate brain functions, in which their final targets may be voltage-gated channels of sodium and potassium.

Soma as a Source of Sequential Spikes at Cortical Pyramidal Neurons

State Key Lab for Brain and Cognitive Sciences, Institute of Biophysics, Chinese Academy of Sciences, Beijing China 100101.The neurons integrate signals from numerous synaptic inputs and produce sequential action potentials as digital codes to carry various messages of controlling well-organized behaviors and cognition. In terms of the source for their initiation, short-time depolarization pulses evoke a single spike at axon hillock. However, action potentials at the neurons in vivo are induced by long-time depolarization. The source for long-duration signals to induce digital spikes needs to be addressed, which we studied at pyramidal neurons in layer 4 of cortical slices by dual-recording on the soma and axonal bleb of identical neurons. In intracellular recording in vivo, the duration of membrane depolarization falls into a range of 60∼1600 ms. These depolarization pulses were injected into the soma and axon of identical cortical pyramidal neurons to induce spikes. Theoretically, the primary location for spikes’ encoding should have more efficient input-output and the highest ability to fire sequential spikes. In whole-cell recording, these physiological signals induce somatic spikes with efficient input-output, the highest spiking ability, lower thresholds and shorter refractory periods, compared with axonal ones. In single-channel recording, voltage-gated sodium channels (VGSC) during a pre-depolarization, which mimics long-time signals, show less inactivation and easier reactivation at the soma than axon. Less inactivated and easily reactivated somatic VGSCs in Neuron model simulate a somatic source of sequential spikes. Based on our data from experiments to theoretical modeling and computational simulation back to experiments, we conclude that physiological inputs primarily trigger the soma to encode neuronal digital signals.

Axons Amplify Somatic Incomplete Spikes into Uniform Amplitudes in Mouse Cortical Pyramidal Neurons

BackgroundAction potentials are the essential unit of neuronal encoding. Somatic sequential spikes in the central nervous system appear various in amplitudes. To be effective neuronal codes, these spikes should be propagated to axonal terminals where they activate the synapses and drive postsynaptic neurons. It remains unclear whether these effective neuronal codes are based on spike timing orders and/or amplitudes.Methodology/Principal FindingsWe investigated this fundamental issue by simultaneously recording the axon versus soma of identical neurons and presynaptic vs. postsynaptic neurons in the cortical slices. The axons enable somatic spikes in low amplitude be enlarged, which activate synaptic transmission in consistent patterns. This facilitation in the propagation of sequential spikes through the axons is mechanistically founded by the short refractory periods, large currents and high opening probability of axonal voltage-gated sodium channels.Conclusion/SignificanceAn amplification of somatic incomplete spikes into axonal complete ones makes sequential spikes to activate consistent synaptic transmission. Therefore, neuronal encoding is likely based on spike timing order, instead of graded analogues.

Intracellular Ca 2+ regulates spike encoding at cortical GABAergic neurons and cerebellar Purkinje cells differently

Spike encoding at GABAergic neurons plays an important role in maintaining the homeostasis of brain functions for well-organized behaviors. The rise of intracellular Ca 2+ in GABAergic neurons causes synaptic plasticity. It is not clear how intracellular Ca 2+ influences their spike encoding. We have investigated this issue at GFP-labeled GABAergic cortical neurons and cerebellar Purkinje cells by whole-cell recording in mouse brain slices. Our results show that an elevation of intracellular Ca 2+ by infusing adenophostin-A lowers spike encoding at GABAergic cortical neurons and enhances encoding ability at cerebellar Purkinje cells. These differential effects of cytoplasmic Ca 2+ on spike encoding are mechanistically associated with Ca 2+-induced changes in the refractory periods and threshold potentials of sequential spikes, as well as with various expression ratios of CaM-KII to calcineurin in GABAergic cortical neurons and cerebellar Purkinje cells.

Axon Amplifies Somatic Sequential Spikes

Action potentials are an essential form of neuronal encoding. Sequential spikes in various amplitudes, to be effective neural codes, should be propagated securely via the axons to activate the synapses and drive postsynaptic neurons. We investigated how the axon propagates sequential spikes by simultaneously recording the axon vs. soma in same cortical neurons. The functionally intact axons enable somatic spikes in lower amplitudes be enlarged, which induce synaptic transmission patterns constantly. This facilitation of the axons to spike propagation is associated with shorter refractory periods of sequential spikes and of voltage-gated sodium channels on the axon vs. soma. Therefore, the axons facilitate the propagation of somatic sequential spikes to presynaptic terminals and set neuronal encoding in spike timing order.