A tool for synthesizing spike trains with realistic interference

Spike detection and spike sorting techniques are often difficult to assess because of the lack of ground truth data (i.e., spike timings for each neuron). This is particularly important for in vitro recordings where the signal to noise ratio is poor (as is the case for multi-electrode arrays at the bottom of a cell culture dish). We present an analysis of the transmission of intracellular signals from neurons to an extracellular electrode, and a set of MATLAB functions based on this analysis. These produce realistic signals from neighboring neurons as well as interference from more distant neurons, and Gaussian noise. They thus generate realistic but controllable synthetic signals (for which the ground truth is known) for assessing spike detection and spike sorting techniques. They can also be used to generate realistic (non-Gaussian) background noise. We use signals generated in this way to compare two automated spike-sorting techniques. The software is available freely on the web.     

How many neurons can we see with current spike sorting algorithms?

Recent studies highlighted the disagreement between the typical number of neurons observed with extracellular recordings and the ones to be expected based on anatomical and physiological considerations. This disagreement has been mainly attributed to the presence of sparsely firing neurons. However, it is also possible that this is due to limitations of the spike sorting algorithms used to process the data. To address this issue, we used realistic simulations of extracellular recordings and found a relatively poor spike sorting performance for simulations containing a large number of neurons. In fact, the number of correctly identified neurons for single-channel recordings showed an asymptotic behavior saturating at about 8-10 units, when up to 20 units were present in the data. This performance was significantly poorer for neurons with low firing rates, as these units were twice more likely to be missed than the ones with high firing rates in simulations containing many neurons. These results uncover one of the main reasons for the relatively low number of neurons found in extracellular recording and also stress the importance of further developments of spike sorting algorithms.     

Computing spike directivity with tetrodes

The ability of neurons to generate electrical signals is strongly dependent on the evolution of ion-specific pumps and channels that allow the transfer of charges under the influence of electric fields and concentration gradients. This paper presents a novel method by which flow of these charge fluxes may be computed to provide directivity of charge movement. Simulations of charge flow as well as actual electrophysiological data recorded by tetrodes are used to demonstrate the method. The propagation of charge fluxes in space in data from simulation and actual recordings during action potential can be analyzed using signals recorded by tetrodes. Variation in spike directivity can be estimated by computing singular value decomposition of the estimated 3D trajectory data. The analysis of the spike model can be accomplished by performing simulations of presumed equivalent moving charges recorded by the tetrode tips. For in vivo spike recordings, the variation of spike directivity could be obtained using several spikes of selected neurons considering the charge movement model (CMM). The relationship between computer simulation results and tetrode data recordings is examined. The paper concludes by showing that the method for calculating directivity in actual spike recordings is robust. The method allows for improved filtering of data and more importantly may shed light on furthering the study of spatio-temporal encoding in neurons.     

Signal-to-noise ratio improvement in multiple electrode recording

Recordings of spike trains made with microwires or silicon electrodes include more noise from various sources that contaminate the observed spike shapes compared with recordings using sharp microelectrodes. This is a particularly serious problem if spike shape sorting is required to separate the several trains that might be observed on a particular electrode. However, if recordings are made with an array of such electrodes, there are several mathematical methods to improve the effective signal (spikes) to noise ratio, thus considerably reducing inaccuracy in spike detection and shape sorting. We compare the theoretical basis of three such methods and evaluate their performance with simulated and real data.     

Detection of neuronal spikes using an adaptive threshold based on the max-min spread sorting method

Neuronal spike information can be used to correlate neuronal activity to various stimuli, to find target neural areas for deep brain stimulation, and to decode intended motor command for brain-machine interface. Typically, spike detection is performed based on the adaptive thresholds determined by running root-mean-square (RMS) value of the signal. Yet conventional detection methods are susceptible to threshold fluctuations caused by neuronal spike intensity. In the present study we propose a novel adaptive threshold based on the max-min spread sorting method. On the basis of microelectrode recording signals and simulated signals with Gaussian noises and colored noises, the novel method had the smallest threshold variations, and similar or better spike detection performance than either the RMS-based method or other improved methods. Moreover, the detection method described in this paper uses the reduced features of raw signal to determine the threshold, thereby giving a simple data manipulation that is beneficial for reducing the computational load when dealing with very large amounts of data (as multi-electrode recordings).     

Development of resurgent and persistent sodium currents in mesencephalic trigeminal neurons

Sodium channels play multiple roles in the formation of neural membrane properties in mesencephalic trigeminal (Mes V) neurons and in other neural systems. Mes V neurons exhibit conditional robust high‐frequency spike discharges. As previously reported, resurgent and persistent sodium currents (I_NaR and I_NaP, respectively) may carry small currents at subthreshold voltages that contribute to generation of spike firing. These currents play an important role in maintaining and allowing high‐frequency spike discharge during a burst. In the present study, we investigated the developmental changes in tetrodotoxin‐sensitive I_NaR and I_NaP underlying high‐frequency spike discharges in Mes V neurons. Whole‐cell patch‐clamp recordings showed that both current densities increased one and a half times from postnatal day (P) 0–6 neurons to P7–14 neurons. Although these neurons do not exhibit subthreshold oscillations or burst discharges with high‐frequency firing, I_NaR and I_NaP do exist in Mes V neurons at P0–6. When the spike frequency at rheobase was examined in firing Mes V neurons, the developmental change in firing frequency among P7–14 neurons was significant. I_NaR and I_NaP density at ?40 mV also increased significantly among P7–14 neurons. The change to an increase in excitability in the P7–14 group could result from this quantitative change in I_NaP. In neurons older than P7 that exhibit repetitive firing, quantitative increases in I_NaR and I_NaP density may be major factors that facilitate and promote high‐frequency firing as a function of age in Mes V neurons.     

Performance comparison of extracellular spike sorting algorithms for single-channel recordings

Proper classification of action potentials from extracellular recordings is essential for making an accurate study of neuronal behavior. Many spike sorting algorithms have been presented in the technical literature. However, no comparative analysis has hitherto been performed. In our study, three widely-used publicly-available spike sorting algorithms (WaveClus, KlustaKwik, OSort) were compared with regard to their parameter settings. The algorithms were evaluated using 112 artificial signals (publicly available online) with 2-9 different neurons and varying noise levels between 0.00 and 0.60. An optimization technique based on Adjusted Mutual Information was employed to find near-optimal parameter settings for a given artificial signal and algorithm. All three algorithms performed significantly better (p < 0.01) with optimized parameters than with the default ones. WaveClus was the most accurate spike sorting algorithm, receiving the best evaluation score for 60% of all signals. OSort operated at almost five times the speed of the other algorithms. In terms of accuracy, OSort performed significantly less well (p < 0.01) than WaveClus for signals with a noise level in the range 0.15-0.30. KlustaKwik achieved similar scores to WaveClus for signals with low noise level 0.00-0.15 and was worse otherwise. In conclusion, none of the three compared algorithms was optimal in general. The accuracy of the algorithms depended on proper choice of the algorithm parameters and also on specific properties of the examined signal.     

New approaches to eliminating common-noise artifacts in recordings from intracortical microelectrode arrays: Inter-electrode correlation and virtual referencing

Intracortical microelectrode arrays record multi-unit extracellular activity for neurophysiology studies and for brain–machine interface applications. The common first step is neural spike-detection; a process complicated by common-noise signals from motion artifacts, electromyographic activity, and electric field pickup, especially in awake/behaving subjects. Often common-noise spikes are very similar to neural spikes in their magnitude, spectral, and temporal features. Provided sufficient spacing exists between electrodes of the array, a local neural spike is rarely recorded on multiple electrodes simultaneously. This is not true for distant common-noise sources. Two new techniques compatible with standard spike-detection schemes are introduced and evaluated. The first method, virtual referencing (VR), takes the average recording from all functional electrodes in the array (represents the signal from a virtual-electrode at the array's center) and subtracts it from the test electrode signal. The second method, inter-electrode correlation (IEC), computes a correlation coefficient between threshold exceeding candidate spike segments on the test electrode and concurrent segments from remaining electrodes. When sufficient correlation is detected, the candidate spike is rejected as originating from a distant common-noise source. The performance of these algorithms was compared with standard thresholding and differential referencing approaches using neural recordings from un-anaesthetized rats. By evaluating characteristics of mean-spike waveforms generated by each method under different levels of common-noise, it was found that IEC consistently offered the most robust means of neural spike-detection. Furthermore, IEC's rejection of supra-threshold events not likely originating from local neurons significantly reduces data handling for downstream spike sorting and processing operations.     

Visualization of multi-neuron activity by simultaneous optimization of clustering and dimension reduction

The recent development of arrays of microelectrodes have enabled simultaneous recordings of the activities of more than 100 neurons. However, it is difficult to visualize activity patterns across many neurons and gain some intuition about issues such as whether the patterns are related to some functions, e.g. perceptual categories. To explore the issues, we used a variational Bayes algorithm to perform clustering and dimension reduction simultaneously. We employed both artificial data and real neuron data to examine the performance of our algorithm. We obtained better clustering results than in a subspace that were obtained by principal component analysis.     

Spike sorting with hidden Markov models

The ability to detect and sort overlapping spike waveforms in extracellular recordings is key to studies of neural coding at high spatial and temporal resolution. Most spike-sorting algorithms are based on initial spike detection (e.g. by a voltage threshold) and subsequent waveform classification. Much effort has been devoted to the clustering step, despite the fact that conservative spike detection is notoriously difficult in low signal-to-noise conditions and often entails many spike misses. Hidden Markov models (HMMs) can serve as generative models for continuous extracellular data records. These models naturally combine the spike detection and classification steps into a single computational procedure. They unify the advantages of independent component analysis (ICA) and overlap-search algorithms because they blindly perform source separation even in cases where several neurons are recorded on a single electrode. We apply HMMs to artificially generated data and to extracellular signals recorded with glass electrodes. We show that in comparison with state-of-art spike-sorting algorithms, HMM-based spike sorting exhibits a comparable number of false positive spike classifications but many fewer spike misses.     

Self‐organization of repetitive spike patterns in developing neuronal networks in vitro

The appearance of spontaneous correlated activity is a fundamental feature of developing neuronal networks in vivo and in vitro. To elucidate whether the ontogeny of correlated activity is paralleled by the appearance of specific spike patterns we used a template‐matching algorithm to detect repetitive spike patterns in multi‐electrode array recordings from cultures of dissociated mouse neocortical neurons between 6 and 15 days in vitro (div). These experiments demonstrated that the number of spiking neurons increased significantly between 6 and 15 div, while a significantly synchronized network activity appeared at 9 div and became the main discharge pattern in the subsequent div. Repetitive spike patterns with a low complexity were first observed at 8 div. The number of repetitive spike patterns in each dataset as well as their complexity and recurrence increased during development in vitro. The number of links between neurons implicated in repetitive spike patterns, as well as their strength, showed a gradual increase during development. About 8% of the spike sequences contributed to more than one repetitive spike patterns and were classified as core patterns. These results demonstrate for the first time that defined neuronal assemblies, as represented by repetitive spike patterns, appear quite early during development in vitro, around the time synchronized network burst become the dominant network pattern. In summary, these findings suggest that dissociated neurons can self‐organize into complex neuronal networks that allow reliable flow and processing of neuronal information already during early phases of development.     

A low-cost single-board solution for real-time, unsupervised waveform classification of multineuron recordings

We describe a low-cost single-board system for unsupervised, real-time spike sorting of recordings from a number of neurons on a single microelectrode. The maximum number of spike classes depends on the quality of the recording; it will typically be between 2 and 5. The spike sorter communicates with a conventional microcomputer through a standard serial port (RS232). For typical firing rates as measured in the mammalian central nervous system, this set-up will accommodate up to some 10 parallel spike sorters for as many separate microelectrodes.     

Failure in identification of overlapping spikes from multiple neuron activity causes artificial correlations

Recording of multiple neurons from a single electrode is common practice during extra-cellular recordings. Separation and sorting of spikes originating from the different neurons can be performed either on-line or off-line using multiple methods for pattern matching. However, all spike sorting techniques fail either fully or partially in identifying spikes from multiple neurons when they overlap due to occurrence within a short time interval. This failure, that we termed the 'shadowing effect', causes the well-known phenomenon of decreased cross-correlation at zero offset. However, the shadowing effect also causes other artifacts in the auto and cross-correlation of the recorded neurons. These artifacts are significant mainly in brain areas with high firing rate or increased firing synchrony leading to a high probability of spike overlap. Cross correlation of cells recorded from the same electrodes tends to reflect the autocorrelation functions of the two cells, even when there are no functional interactions between the cells. Therefore, the cross-correlation function tends to have a short-term (about the length of the refractory period) peak. A long-term (hundreds of milliseconds to a few seconds) trough in the cross-correlation can be seen in cells with bursting and pausing activities recorded from the same electrode. Even the autocorrelation functions of the recorded neurons feature firing properties of other neurons recorded from the same electrode. Examples of these effects are given from our recordings in the globus pallidus of behaving primates and from the literature. Results of simulations of independent simple model neurons exhibit the same properties as the recorded neurons. The effect is analyzed and can be estimated to enable better evaluation of the underlying firing patterns and the actual synchronization of neighboring neurons recorded by a single electrode.     

An improved method for the estimation of firing rate dynamics using an optimal digital filter

In most neural systems, neurons communicate by means of sequences of action potentials or 'spikes'. Information encoded by spike trains is often quantified in terms of the firing rate which emphasizes the frequency of occurrence of action potentials rather than their exact timing. Common methods for estimating firing rates include the rate histogram, the reciprocal interspike interval, and the spike density function. In this study, we demonstrate the limitations of these aforementioned techniques and propose a simple yet more robust alternative. By convolving the spike train with an optimally designed Kaiser window, we show that more robust estimates of firing rate are obtained for both low and high-frequency inputs. We illustrate our approach by considering spike trains generated by simulated as well as experimental data obtained from single-unit recordings of first-order sensory neurons in the vestibular system. Improvements were seen in the prevention of aliasing, phase and amplitude distortion, as well as in the noise reduction for sinusoidal and more complex input profiles. We review the generality of the approach, and show that it can be adapted to describe neurons with sensory or motor responses that are characterized by marked nonlinearities. We conclude that our method permits more robust estimates of neural dynamics than conventional techniques across all stimulus conditions.     

Electrophysiological evidence on epileptiform activity enhanced by electrical stimulation of teeth in rats

Abstract

The aim of this study was to evaluate the possible effects of electrical stimulation (ES) of tooth on penicillin-induced epileptiform activity in rats.

Experiment was realized on 24 adult male Sprague Dawley rats. Rats were assigned three groups [stimulation group (SG), penicillin group (PG), and penicillin+stimulation group (PSG)]. In SG, ES was only applied. Ten pulses of electrical current were delivered to the teeth for a duration of 2 milliseconds at 1-second intervals from a stimulator. Currents were applied in the range of 40–240 μA with 40 μA steps. Electrocorticography (ECoG) recordings were taken before and after ES. In PG, ECoG recordings were taken before and during the injection of penicillin. In PSG, after epileptiform activity was induced, ES was applied and ECoG recordings were taken as in SG. All the data were analyzed with Student’s t test. Applied currents did not cause any epileptiform activity in SG. When the PSG was compared with the PG it was seen that the spike frequency of epileptiform activity increased in a statistically significant way after application of 240 μA (P < 0·05). On the other hand current application caused an increase in the spike amplitude of the PSG compared with the amplitude of the PG, but it was not statistically significant.

We concluded that ES of tooth with high current can trigger epileptiform activity in rats. For this reason, further research is required to evaluate the effects of ES of tooth for pulp testing on epileptic human subjects and antiepileptic drug users.     

Activity dynamics and behavioral correlates of CA3 and CA1 hippocampal pyramidal neurons

The CA3 and CA1 pyramidal neurons are the major principal cell types of the hippocampus proper. The strongly recurrent collateral system of CA3 cells and the largely parallel‐organized CA1 neurons suggest that these regions perform distinct computations. However, a comprehensive comparison between CA1 and CA3 pyramidal cells in terms of firing properties, network dynamics, and behavioral correlations is sparse in the intact animal. We performed large‐scale recordings in the dorsal hippocampus of rats to quantify the similarities and differences between CA1 (n > 3,600) and CA3 (n > 2,200) pyramidal cells during sleep and exploration in multiple environments. CA1 and CA3 neurons differed significantly in firing rates, spike burst propensity, spike entrainment by the theta rhythm, and other aspects of spiking dynamics in a brain state‐dependent manner. A smaller proportion of CA3 than CA1 cells displayed prominent place fields, but place fields of CA3 neurons were more compact, more stable, and carried more spatial information per spike than those of CA1 pyramidal cells. Several other features of the two cell types were specific to the testing environment. CA3 neurons showed less pronounced phase precession and a weaker position versus spike‐phase relationship than CA1 cells. Our findings suggest that these distinct activity dynamics of CA1 and CA3 pyramidal cells support their distinct computational roles. © 2012 Wiley Periodicals, Inc.     

Embedded ensemble encoding hypothesis: The role of the “Prepared” cell

We here reconsider current theories of neural ensembles in the context of recent discoveries about neuronal dendritic physiology. The key physiological observation is that the dendritic plateau potential produces sustained depolarization of the cell body (amplitude 10–20 mV, duration 200–500 ms). Our central hypothesis is that synaptically‐evoked dendritic plateau potentials lead to a prepared state of a neuron that favors spike generation. The plateau both depolarizes the cell toward spike threshold, and provides faster response to inputs through a shortened membrane time constant. As a result, the speed of synaptic‐to‐action potential (AP) transfer is faster during the plateau phase. Our hypothesis relates the changes from “resting” to “depolarized” neuronal state to changes in ensemble dynamics and in network information flow. The plateau provides the Prepared state (sustained depolarization of the cell body) with a time window of 200–500 ms. During this time, a neuron can tune into ongoing network activity and synchronize spiking with other neurons to provide a coordinated Active state (robust firing of somatic APs), which would permit “binding” of signals through coordination of neural activity across a population. The transient Active ensemble of neurons is embedded in the longer‐lasting Prepared ensemble of neurons. We hypothesize that “embedded ensemble encoding” may be an important organizing principle in networks of neurons.     

Spike shape analysis of electromyography for parkinsonian tremor evaluation

Introduction: Standard electromyography (EMG) parameters have limited utility for evaluation of Parkinson disease (PD) tremor. Spike shape analysis (SSA) EMG parameters are more sensitive than standard EMG parameters for studying motor control mechanisms in healthy subjects. SSA of EMG has not been used to assess parkinsonian tremor. This study assessed the utility of SSA and standard time and frequency analysis for electromyographic evaluation of PD‐related resting tremor. Methods: We analyzed 1‐s periods of EMG recordings to detect nontremor and tremor signals in relaxed biceps brachii muscle of seven mild to moderate PD patients. Results: SSA revealed higher mean spike amplitude, duration, and slope and lower mean spike frequency in tremor signals than in nontremor signals. Standard EMG parameters (root mean square, median, and mean frequency) did not show differences between the tremor and nontremor signals. Conclusions: SSA of EMG data is a sensitive method for parkinsonian tremor evaluation. Muscle Nerve 52 : 1096–1098, 2015     

Characterizing synaptic conductance fluctuations in cortical neurons and their influence on spike generation

Cortical neurons are subject to sustained and irregular synaptic activity which causes important fluctuations of the membrane potential (V(m)). We review here different methods to characterize this activity and its impact on spike generation. The simplified, fluctuating point-conductance model of synaptic activity provides the starting point of a variety of methods for the analysis of intracellular V(m) recordings. In this model, the synaptic excitatory and inhibitory conductances are described by Gaussian-distributed stochastic variables, or "colored conductance noise". The matching of experimentally recorded V(m) distributions to an invertible theoretical expression derived from the model allows the extraction of parameters characterizing the synaptic conductance distributions. This analysis can be complemented by the matching of experimental V(m) power spectral densities (PSDs) to a theoretical template, even though the unexpected scaling properties of experimental PSDs limit the precision of this latter approach. Building on this stochastic characterization of synaptic activity, we also propose methods to qualitatively and quantitatively evaluate spike-triggered averages of synaptic time-courses preceding spikes. This analysis points to an essential role for synaptic conductance variance in determining spike times. The presented methods are evaluated using controlled conductance injection in cortical neurons in vitro with the dynamic-clamp technique. We review their applications to the analysis of in vivo intracellular recordings in cat association cortex, which suggest a predominant role for inhibition in determining both sub- and supra-threshold dynamics of cortical neurons embedded in active networks.     

Putative excitatory and putative inhibitory inputs are localised in different dendritic domains in a Drosophila flight motoneuron

Input–output computations of individual neurons may be affected by the three‐dimensional structure of their dendrites and by the location of input synapses on specific parts of their dendrites. However, only a few examples exist of dendritic architecture which can be related to behaviorally relevant computations of a neuron. By combining genetic, immunohistochemical and confocal laser scanning methods this study estimates the location of the spike‐initiating zone and the dendritic distribution patterns of putative synaptic inputs on an individually identified Drosophila flight motorneuron, MN5. MN5 is a monopolar neuron with > 4000 dendritic branches. The site of spike initiation was estimated by mapping sodium channel immunolabel onto geometric reconstructions of MN5. Maps of putative excitatory cholinergic and of putative inhibitory GABAergic inputs on MN5 dendrites were created by charting tagged Dα7 nicotinic acetylcholine receptors and Rdl GABA_A receptors onto MN5 dendritic surface reconstructions. Although these methods provide only an estimate of putative input synapse distributions, the data indicate that inhibitory and excitatory synapses were located preferentially on different dendritic domains of MN5 and, thus, computed mostly separately. Most putative inhibitory inputs were close to spike initiation, which was consistent with sharp inhibition, as predicted previously based on recordings of motoneuron firing patterns during flight. By contrast, highest densities of putative excitatory inputs at more distant dendritic regions were consistent with the prediction that, in response to different power demands during flight, tonic excitatory drive to flight motoneuron dendrites must be smoothly translated into different tonic firing frequencies.