Auditory brainstem development of naked mole-rats (Heterocephalus glaber).

Life underground often leads to animals having specialized auditory systems to accommodate the constraints of acoustic transmission in tunnels. Despite living underground, naked mole-rats use a highly vocal communication system, implying that they rely on central auditory processing. However, little is known about these animals' central auditory system, and whether it follows a similar developmental time course as other rodents. Naked mole-rats show slowed development in the hippocampus suggesting they have altered brain development compared to other rodents. Here, we measured morphological characteristics and voltage-gated potassium channel Kv3.3 expression and protein levels at different key developmental time points (postnatal days 9, 14, 21 and adulthood) to determine whether the auditory brainstem (lateral superior olive and medial nucleus of the trapezoid body) develops similarly to two common auditory rodent model species: gerbils and mice. Additionally, we measured the hearing onset of naked mole-rats using auditory brainstem response recordings at the same developmental timepoints. In contrast with other work in naked mole-rats showing that they are highly divergent in many aspects of their physiology, we show that naked mole-rats have a similar hearing onset, between postnatal day (P) 9 and P14, to many other rodents. On the other hand, we show some developmental differences, such as a unique morphology and Kv3.3 protein levels in the brainstem.

Mechanisms underlying auditory processing deficits in Fragile X syndrome.

Autism spectrum disorders (ASD) are strongly associated with auditory hypersensitivity or hyperacusis (difficulty tolerating sounds). Fragile X syndrome (FXS), the most common monogenetic cause of ASD, has emerged as a powerful gateway for exploring underlying mechanisms of hyperacusis and auditory dysfunction in ASD. This review discusses examples of disruption of the auditory pathways in FXS at molecular, synaptic, and circuit levels in animal models as well as in FXS individuals. These examples highlight the involvement of multiple mechanisms, from aberrant synaptic development and ion channel deregulation of auditory brainstem circuits, to impaired neuronal plasticity and network hyperexcitability in the auditory cortex. Though a relatively new area of research, recent discoveries have increased interest in auditory dysfunction and mechanisms underlying hyperacusis in this disorder. This rapidly growing body of data has yielded novel research directions addressing critical questions regarding the timing and possible outcomes of human therapies for auditory dysfunction in ASD.

Enhancement of de novo sequencing, assembly and annotation of the Mongolian gerbil genome with transcriptome sequencing and assembly from several different tissues.

BACKGROUND: The Mongolian gerbil (Meriones unguiculatus) has historically been used as a model organism for the auditory and visual systems, stroke/ischemia, epilepsy and aging related research since 1935 when laboratory gerbils were separated from their wild counterparts. In this study we report genome sequencing, assembly, and annotation further supported by transcriptome sequencing and assembly from 27 different tissues samples. RESULTS: The genome was sequenced using Illumina HiSeq 2000 and after assembly resulted in a final genome size of 2.54 Gbp with contig and scaffold N50 values of 31.4 Kbp and 500.0 Kbp, respectively. Based on the k-mer estimated genome size of 2.48 Gbp, the assembly appears to be complete. The genome annotation was supported by transcriptome data that identified 31,769 (> 2000 bp) predicted protein-coding genes across 27 tissue samples. A BUSCO search of 3023 mammalian groups resulted in 86% of curated single copy orthologs present among predicted genes, indicating a high level of completeness of the genome. CONCLUSIONS: We report the first de novo assembly of the Mongolian gerbil genome enhanced by assembly of transcriptome data from several tissues. Sequencing of this genome and transcriptome increases the utility of the gerbil as a model organism, opening the availability of now widely used genetic tools.

A recording chamber for small volume slice electrophysiology.

Electrophysiological recordings from brain slices are typically performed in small recording chambers that allow for the superfusion of the tissue with artificial extracellular solution (ECS), while the chamber holding the tissue is mounted in the optical path of a microscope to image neurons in the tissue. ECS itself is inexpensive, and thus superfusion rates and volumes of ECS consumed during an experiment using standard ECS are not critical. However, some experiments require the addition of expensive pharmacological agents or other chemical compounds to the ECS, creating a need to build superfusion systems that operate on small volumes while still delivering appropriate amounts of oxygen and other nutrients to the tissue. We developed a closed circulation tissue chamber for slice recordings that operates with small volumes of bath solution in the range of 1.0 to 2.6 ml and a constant oxygen/carbon dioxide delivery to the solution in the bath. In our chamber, the ECS is oxygenated and recirculated directly in the recording chamber, eliminating the need for tubes and external bottles/containers to recirculate and bubble ECS and greatly reducing the total ECS volume required for superfusion. At the same time, the efficiency of tissue oxygenation and health of the section are comparable to standard superfusion methods. We also determined that the small volume of ECS contains a sufficient amount of nutrients to support the health of a standard brain slice for several hours without concern for either depletion of nutrients or accumulation of waste products.

Glycinergic inhibition to the medial nucleus of the trapezoid body shows prominent facilitation and can sustain high levels of ongoing activity.

Neurons in the medial nucleus of the trapezoid body (MNTB) are well known for their prominent excitatory inputs, mediated by the calyx of Held. Less attention has been paid to the prominent inhibitory inputs that MNTB neurons also receive. Because of their auditory nature, both excitatory and inhibitory synapses are highly active in vivo. These high levels of activity are known to reduce excitatory synaptic currents considerably, such that in vivo synaptic currents produced by the calyx are smaller than typically measured in standard brain slice experiments. The goal of this study was to investigate the properties of the inhibitory inputs in the Mongolian gerbil (Meriones unguiculatus) under activity levels that correspond to those in the intact brain to facilitate a direct comparison between the two inputs. Our results suggest that inhibitory inputs to MNTB are largely mediated by a fast and phasic glycinergic component, and to a lesser degree by a GABAergic component. The glycinergic component can sustain prolonged high levels of activity. Even when challenged with stimulus patterns consisting of thousands of stimuli over tens of minutes, glycinergic inputs to MNTB maintain large conductances and fast decays and even facilitate substantially when the stimulation frequency is increased. The inhibition is mediated by a relatively small number of independent input fibers. The data presented here suggest that inhibitory inputs to MNTB sustain high levels of activity and need to be considered for a full understanding of mechanisms underlying processing of auditory information in MNTB.

A fully automatic multichannel neural spike sorting algorithm with spike reduction and positional feature.

Objective: The sorting of neural spike data recorded by multichannel and high channel neural probes such as Neuropixels, especially in real-time, remains a significant technical challenge. Most neural spike sorting algorithms focus on sorting neural spikes post-hoc for high sorting accuracy-but reducing the processing delay for fast sorting, potentially even live sorting, is generally not possible with these algorithms.Approach: Here we report our Graph nEtwork Multichannel sorting (GEMsort) algorithm, which is largely based on graph network, to allow rapid neural spike sorting for multiple neural recording channels. This was accomplished by two innovations: In GEMsort, duplicated neural spikes recorded from multiple channels were eliminated from duplicate channels by only selecting the highest amplitude neural spike in any channel for subsequent processing. In addition, the channel from which the representative neural spike was recorded was used as an additional feature to differentiate between neural spikes recorded from different neurons having similar temporal features.Main results: Synthetic and experimentally recorded multichannel neural recordings were used to evaluate the sorting performance of GEMsort. The sorting results of GEMsort were also compared with two other state-of-the-art sorting algorithms (Kilosort and Mountainsort) in sorting time and sorting agreements.Significance: GEMsort allows rapidly sort neural spikes and is highly suitable to be implemented with digital circuitry for high processing speed and channel scalability.

Degradation in Binaural and Spatial Hearing and Auditory Temporal Processing Abilities as a Function of Aging.

Objective: Sensorineural hearing loss is common with advancing age, but even with normal or near normal hearing in older persons, performance deficits are often seen for suprathreshold listening tasks such as understanding speech in background noise or localizing sound direction. This suggests there is also a more central source of the problem. Objectives of this study were to examine as a function of age (young adult to septuagenarian) performance on: 1) a spatial acuity task examining lateralization ability, and a spatial speech-in-noise (SSIN) recognition task, both measured in a hemi-anechoic sound field using a circular horizontal-plane loudspeaker array, and 2) a suprathreshold auditory temporal processing task and a spectro-temporal processing task, both measured under headphones. Further, we examined any correlations between the measures.DesignSubjects were 48 adults, aged 21 to 78, with either normal hearing or only a mild sensorineural hearing loss through 4000 Hz. The lateralization task measured minimum audible angle (MAA) for 500 and 4000 Hz narrowband noise (NBN) bursts in diffuse background noise for both an on-axis (subject facing 0°) and off-axis (facing 45°) listening condition at signal-to-noise ratios (SNRs) of -3, -6, -9, and -12 dB. For 42 of the subjects, SSIN testing was also completed for key word recognition in sentences in multi-talker babble noise; specifically, the separation between speech and noise loudspeakers was adaptively varied to determine the difference needed for 40% and 80% correct performance levels. Finally, auditory temporal processing ability was examined using the Temporal Fine Structure test (44 subjects), and the Spectro-Temporal Modulation test (46 subjects). Results: Mean lateralization performances were poorer (larger MAAs) in older compared to younger subjects, particularly in the more adverse listening conditions (4000 Hz, off-axis, and poorer SNRs). Performance variability was notably higher for older subjects than for young adults. The 4000 Hz NBN bursts produced larger MAAs than did 500 Hz NBN bursts. The SSIN data also showed declining mean performance with age at both criterion levels, with greater variability again found for older subjects. Spearman rho analyses revealed some low to moderate, but significant correlation coefficients for age versus MAA and age versus SSIN results. A low but significant correlation was also observed between the most adverse MAA and SSIN conditions. Results from both the TFS and STM assessments showed decreased mean performance with aging, and revealed moderate, significant correlations, with the strongest relationship shown with the TFS test. Finally, of note, extended-high-frequency (EHF) hearing loss (measured between 9000 and 16,000 Hz) was found in older but not young subjects, and correlated with decreasing performance on several tasks. Conclusions: Particularly for more adverse listening conditions, age-related deficits were found on both of the spatial hearing tasks and in temporal and spectro-temporal processing abilities. It may be that deficits in temporal processing ability contribute to poorer spatial hearing performance in older subjects due to inaccurate coding of binaural/interaural timing information sent from the periphery to the brainstem. In addition, EHF hearing loss may be a coexisting factor in the reduced performance in older subjects.

Computational model for synthesizing auditory brainstem responses to assess neuronal alterations in aging and autistic animal models.

The auditory brainstem response (ABR) is a widely used objective electrophysiology measure for non-invasively assessing auditory function and neural activities in the auditory brainstem, but its ability to reflect detailed neuronal processes is limited due to the averaging nature of the electroencephalogram recordings. This study addresses this limitation by developing a computational model of the auditory brainstem which is capable of synthesizing ABR traces based on a large, population scale neural extrapolation of a spiking neuronal network of auditory brainstem neural circuitry. The model was able to recapitulate alterations in ABR waveform morphology that have been shown to be present in two medical conditions: animal models of autism and aging. Moreover, in both of these conditions, these ABR alterations are caused by known distinct changes in auditory brainstem physiology, and the model could recapitulate these changes. In the autism model, the simulation revealed myelin deficits and hyperexcitability, which caused a decreased wave III amplitude and a prolonged wave III-V interval, consistent with experimentally recorded ABRs in Fmr1-KO mice. In the aging model, the model recapitulated ABRs recorded in aged gerbils and indicated a reduction in activity in the medial nucleus of the trapezoid body (MNTB), a finding validated by confocal imaging data. These results demonstrate not only the model's accuracy but also its capability of linking features of ABR morphologies to underlying neuronal properties and suggesting follow-up physiological experiments. Significance Statement: This study presents a novel computational model of the auditory brainstem, capable of synthesizing auditory brainstem response (ABR) traces by simulating large-scale neuronal activities. Addressing limitations of traditional ABR measurements, the model links ABR waveform features to underlying neuronal properties. Validated using empirical ABRs from animal models of autism and aging, the model accurately reproduced observed ABR alterations, revealing influences of myelin deficits and hyperexcitability in Fragile X syndrome, and degraded inhibitory activity in aging. These findings, supported by experimental data, demonstrate the model's potential for predicting changes in auditory brainstem physiology and guiding further physiological investigations, thus advancing our understanding of auditory neural processes.

Sapphire-Based Optrode for Low Noise Neural Recording and Optogenetic Manipulation.

Electrophysiological recordings of neurons in deep brain regions using optogenetic stimulation are essential to understanding and regulating the role of complex neural activity in biological behavior and cognitive function. Optogenetic techniques have significantly advanced neuroscience research by enabling the optical manipulation of neural activities. Because of the significance of the technique, constant advancements in implantable optrodes that integrate optical stimulation with low-noise, large-scale electrophysiological recording are in demand to improve the spatiotemporal resolution for various experimental designs and future clinical applications. However, robust and easy-to-use neural optrodes that integrate neural recording arrays with high-intensity light emitting diodes (LEDs) are still lacking. Here, we propose a neural optrode based on Gallium Nitride (GaN) on sapphire technology, which integrates a high-intensity blue LED with a 5x2 recording array monolithically for simultaneous neural recording and optogenetic manipulation. To reduce the noise interference between the recording electrodes and the LED, which is in close physical proximity, three metal grounding interlayers were incorporated within the optrode, and their ability to reduce LED-induced artifacts during neural recording was confirmed through both electromagnetic simulations and experimental demonstrations. The capability of the sapphire optrode to record action potentials has been demonstrated by recording the firing of mitral/tuft cells in the olfactory bulbs of mice in vivo. Additionally, the elevation of action potential firing due to optogenetic stimulation observed using the sapphire probe in medial superior olive (MSO) neurons of the gerbil auditory brainstem confirms the capability of this sapphire optrode to precisely access neural activities in deep brain regions under complex experimental designs.

How do short-term changes at synapses fine-tune information processing?

Synaptic transmission is highly dependent on recent activity and can lead to depression or facilitation of synaptic strength. This phenomenon is called "short-term synaptic plasticity" and is shown at all synapses. While much work has been done to understand the mechanisms of short-term changes in the state of synapses, short-term plasticity is often thought of as a mechanistic consequence of the design of a synapse. This review will attempt to go beyond this view and discuss how, on one hand, complex neuronal activity affects the short-term state of synapses, but also how these dynamic changes in synaptic strength affect information processing in return.

Modulation of synaptic input by GABAB receptors improves coincidence detection for computation of sound location.

Interaural time disparities (ITDs) are the primary cues for localisation of low-frequency sound stimuli. ITDs are computed by coincidence-detecting neurones in the medial superior olive (MSO) in mammals. Several previous studies suggest that control of synaptic gain is essential for maintaining ITD selectivity as stimulus intensity increases. Using acute brain slices from postnatal day 7 to 24 (P7­P24) Mongolian gerbils, we confirm that activation of GABAB receptors reduces the amplitude of excitatory and inhibitory synaptic currents to the MSO and, moreover, show that the decay kinetics of IPSCs are slowed in mature animals. During repetitive stimuli, activation of GABAB receptors reduced the amount of depression observed, while PSC suppression and the slowed kinetics were maintained. Additionally, we used physiological and modelling approaches to test the potential impact of GABAB activation on ITD encoding in MSO neurones. Current clamp recordings from MSO neurones were made while pharmacologically isolated excitatory inputs were bilaterally stimulated using pulse trains that simulate ITDs in vitro. MSO neurones showed strong selectivity for bilateral delays. Application of both GABAB agonists and antagonists demonstrate that GABAB modulation of synaptic input can sharpen ITD selectivity. We confirmed and extended these results in a computational model that allowed for independent manipulation of each GABAB-dependent effect. Modelling suggests that modulation of both amplitude and kinetics of synaptic inputs by GABAB receptors can improve precision of ITD computation. Our studies suggest that in vivo modulation of synaptic input by GABAB receptors may act to preserve ITD selectivity across various stimulus conditions.

Coherent Anti-Stokes Raman Spectroscopy (CARS) Application for Imaging Myelination in Brain Slices.

Coherent anti-Stokes Raman spectroscopy (CARS) is a technique classically employed by chemists and physicists to produce a coherent signal of signature vibrations of molecules. However, these vibrational signatures are also characteristic of molecules within anatomical tissue such as the brain, making it increasingly useful and applicable for Neuroscience applications. For example, CARS can measure lipids by specifically exciting chemical bonds within these molecules, allowing for quantification of different aspects of tissue, such as myelin involved in neurotransmission. In addition, compared to other techniques typically used to quantify myelin, CARS can also be set up to be compatible with immunofluorescent techniques, allowing for co-labeling with other markers such as sodium channels or other components of synaptic transmission. Myelination changes are an inherently important mechanism in demyelinating diseases such as multiple sclerosis or other neurological conditions such as Fragile X Syndrome or autism spectrum disorders is an emerging area of research. In conclusion, CARS can be utilized in innovative ways to answer pressing questions in Neuroscience and provide evidence for underlying mechanisms related to many different neurological conditions.

Predicting the Influence of Axon Myelination on Sound Localization Precision Using a Spiking Neural Network Model of Auditory Brainstem.

Spatial hearing allows animals to rapidly detect and localize auditory events in the surrounding environment. The auditory brainstem plays a central role in processing and extracting binaural spatial cues through microsecond-precise binaural integration, especially for detecting interaural time differences (ITDs) of low-frequency sounds at the medial superior olive (MSO). A series of mechanisms exist in the underlying neural circuits for preserving accurate action potential timing across multiple fibers, synapses and nuclei along this pathway. One of these is the myelination of afferent fibers that ensures reliable and temporally precise action potential propagation in the axon. There are several reports of fine-tuned myelination patterns in the MSO circuit, but how specifically myelination influences the precision of sound localization remains incompletely understood. Here we present a spiking neural network (SNN) model of the Mongolian gerbil auditory brainstem with myelinated axons to investigate whether different axon myelination thicknesses alter the sound localization process. Our model demonstrates that axon myelin thickness along the contralateral pathways can substantially modulate ITD detection. Furthermore, optimal ITD sensitivity is reached when the MSO receives contralateral inhibition via thicker myelinated axons compared to contralateral excitation, a result that is consistent with previously reported experimental observations. Our results suggest specific roles of axon myelination for extracting temporal dynamics in ITD decoding, especially in the pathway of the contralateral inhibition.

Characterization of a Jumping Stereotypy in Gerbils (Meriones unguiculatus) and Assessment of Opaque Tubing Enrichment on Stereotypies and Breeding.

Mongolian gerbils can develop stereotypic behaviors, including corner digging. At our institution, gerbils also engage in repetitive corner jumping, which we sought to characterize as a potentially novel stereotypy in gerbils. We then attempted to mitigate this behavior by mimicking the natural habitat by adding intracage environmental complexity. Seventeen gerbil breeding pairs were video recorded in their home cages during the light cycle. Repetitive corner jumping and digging were compared between different times of day to assess when the behaviors occurred and whether they were temporally associated. To determine whether we could reduce the incidence of stereotypic behaviors, we tested a straight tube or 1 of 3 angled opaque tubes in different orientations, which were fitted to the gerbils' preexisting opaque nesting box. Behavior was assessed at baseline and at 1, 4, 8, and 12 wk to evaluate opaque tube placement as an intervention. In addition, breeding efficiency, valuated as the number of gerbil pups born and weaned per breeder pair, was compared with pre- and poststudy data. The number of corner jumps was highest at the end of the light cycle and the majority were associated with corner digging. After placement of the enrichment tubes, an initial increase in corner digging behavior was observed and persisted throughout the study period. The opaque tubes were not associated with significant changes in corner jumping. After adjusting for age, the addition of opaque tubing to gerbil breeding cages was not associated with significant changes in breeding efficiency. The addition of opaque tubing did not effectively address concerns about stereotypic behaviors and was associated with a chronic increase in stereotypic corner digging among breeding gerbil pairs.

Myelination Deficits in the Auditory Brainstem of a Mouse Model of Fragile X Syndrome.

Auditory symptoms are one of the most frequent sensory issues described in people with Fragile X Syndrome (FXS), the most common genetic form of intellectual disability. However, the mechanisms that lead to these symptoms are under explored. In this study, we examined whether there are defects in myelination in the auditory brainstem circuitry. Specifically, we studied myelinated fibers that terminate in the Calyx of Held, which encode temporally precise sound arrival time, and are some of the most heavily myelinated axons in the brain. We measured anatomical myelination characteristics using coherent anti-stokes Raman spectroscopy (CARS) and electron microscopy (EM) in a FXS mouse model in the medial nucleus of the trapezoid body (MNTB) where the Calyx of Held synapses. We measured number of mature oligodendrocytes (OL) and oligodendrocyte precursor cells (OPCs) to determine if changes in myelination were due to changes in the number of myelinating or immature glial cells. The two microscopy techniques (EM and CARS) showed a decrease in fiber diameter in FXS mice. Additionally, EM results indicated reductions in myelin thickness and axon diameter, and an increase in g-ratio, a measure of structural and functional myelination. Lastly, we showed an increase in both OL and OPCs in MNTB sections of FXS mice suggesting that the myelination phenotype is not due to an overall decrease in number of myelinating OLs. This is the first study to show that a myelination defects in the auditory brainstem that may underly auditory phenotypes in FXS.

Axonal Conduction Delay Shapes the Precision of the Spatial Hearing in A Spiking Neural Network Model of Auditory Brainstem.

One method by which the mammalian sound localization pathway localizes sound sources is by analyzing the microsecond-level difference between the arrival times of a sound at the two ears. However, how the neural circuits in the auditory brainstem precisely integrate signals from the two ears, and what the underlying mechanisms are, remains to be understood. Recent studies have reported that variations of axon myelination in the auditory brainstem produces various axonal conduction velocities and sophisticated temporal dynamics, which have not been well characterized in most existing models of sound localization circuits. Here, we present a spiking neural network model of the auditory brainstem to investigate how axon myelinations affect the precision of sound localization. Sound waves with different interaural time differences (ITDs) are encoded and used as stimuli, and the axon properties in the network are adjusted, and the corresponding axonal conduction delays are computed with a multi-compartment axon model. Through the simulation, the sensitivity of ITD perception varies with the myelin thickness of axons in the contralateral input pathways to the medial superior olive (MSO). The ITD perception becomes more precise when the contralateral inhibitory input propagates faster than the contralateral excitatory input. These results indicate that axon myelination and contralateral spike timing influence spatial hearing perception.

Robotic stereotaxic system based on 3D skull reconstruction to improve surgical accuracy and speed.

BACKGROUND: Some experimental approaches in neuroscience research require the precise placement of a recording electrode, pipette or other tool into a specific brain area that can be quite small and/or located deep beneath the surface. This process is typically aided with stereotaxic methods but remains challenging due to a lack of advanced technology to aid the experimenter. Currently, procedures require a significant amount of skill, have a high failure rate, and take up a significant amount of time. NEW METHOD: We developed a next generation robotic stereotaxic platform for small rodents by combining a three-dimensional (3D) skull profiler sub-system and a full six degree-of-freedom (6DOF) robotic platform. The 3D skull profiler is based on structured illumination in which a series of horizontal and vertical line patterns are projected onto an animal skull. These patterns are captured by two two-dimensional (2D) CCD cameras which reconstruct an accurate 3D skull surface profile based on structured illumination and geometrical triangulation. Using the reconstructed 3D profile, the skull can be repositioned using a 6DOF robotic platform to accurately align a surgical tool. RESULTS: The system was evaluated using mechanical measurement techniques, and the accuracy of the platform was demonstrated using agar brain phantoms and animal skulls. Additionally, a small and deep brain nucleus (the medial nucleus of the trapezoid body) were targeted in rodents to confirm the targeting accuracy. CONCLUSIONS: The new stereotaxic system can accomplish "skull-flat" rapidly and precisely and with minimal user intervention, and thus reduces the failure rate of such experiments.

Characterization of Auditory and Binaural Spatial Hearing in a Fragile X Syndrome Mouse Model.

The auditory brainstem compares sound-evoked excitation and inhibition from both ears to compute sound source location and determine spatial acuity. Although alterations to the anatomy and physiology of the auditory brainstem have been demonstrated in fragile X syndrome (FXS), it is not known whether these changes cause spatial acuity deficits in FXS. To test the hypothesis that FXS-related alterations to brainstem circuits impair spatial hearing abilities, a reflexive prepulse inhibition (PPI) task, with variations in sound (gap, location, masking) as the prepulse stimulus, was used on Fmr1 knock-out mice and B6 controls. Specifically, Fmr1 mice show decreased PPI compared with wild-type mice during gap detection, changes in sound source location, and spatial release from masking with no alteration to their overall startle thresholds compared with wild-type mice. Last, Fmr1 mice have increased latency to respond in these tasks, suggesting additional impairments in the pathway responsible for reacting to a startling sound. This study further supports data in humans with FXS that show similar deficits in PPI.

Computationally inexpensive enhanced growing neural gas algorithm for real-time adaptive neural spike clustering.

OBJECTIVE: Real-time closed-loop neural feedback control requires the analysis of action potential traces within several milliseconds after they have been recorded from the brain. The current generation of spike clustering algorithms were mostly designed for off-line use and also require a significant amount of computational resources. A new spike clustering algorithm, termed 'enhanced growing neural gas (EGNG)', was therefore developed that is computationally lightweight and memory conserving. The EGNG algorithm can adapt to changes of the electrophysiological recording environment and can classify both pre-recorded and streaming action potentials. APPROACH: The algorithm only uses a small number of EGNG nodes and edges to learn the neural spike distributions which eliminates the need of retaining the neural data in the system memory to conserve computational resources. Most of the computations revolve around calculating Euclidian distances, which is computationally inexpensive and can be implemented in parallel using digital circuit technology. MAIN RESULTS: EGNG was evaluated off-line using both synthetic and pre-recorded neural spikes. Streaming synthetic neural spikes were also used to evaluate the ability of EGNG to classify action potentials in real-time. The algorithm was also implemented in hardware with a Field Programming Gate Array (FPGA) chip, and the worst-case clustering latency was 3.10 µs, allowing a minimum of 322 580 neural spikes to be clustered per second. SIGNIFICANCE: The EGNG algorithm provides a viable solution to classification of neural spikes in real-time and can be implemented with limited computational resources as a front-end spike clustering unit for future tethered-free and miniaturized closed-loop neural feedback systems.

Low-latency single channel real-time neural spike sorting system based on template matching.

Recent technical advancements in neural engineering allow for precise recording and control of neural circuits simultaneously, opening up new opportunities for closed-loop neural control. In this work, a rapid spike sorting system was developed based on template matching to rapidly calculate instantaneous firing rates for each neuron in a multi-unit extracellular recording setting. Cluster templates were first generated by a desktop computer using a non-parameter spike sorting algorithm (Super-paramagnetic clustering) and then transferred to a field-programmable gate array digital circuit for rapid sorting through template matching. Two different matching techniques-Euclidean distance (ED) and correlational matching (CM)-were compared for the accuracy of sorting and the performance of calculating firing rates. The performance of the system was first verified using publicly available artificial data and was further confirmed with pre-recorded neural spikes from an anesthetized Mongolian gerbil. Real-time recording and sorting from an awake mouse were also conducted to confirm the system performance in a typical behavioral neuroscience experimental setting. Experimental results indicated that high sorting accuracies were achieved for both template-matching methods, but CM can better handle spikes with non-Gaussian spike distributions, making it more robust for in vivo recording. The technique was also compared to several other off-line spike sorting algorithms and the results indicated that the sorting accuracy is comparable but sorting time is significantly shorter than these other techniques. A low sorting latency of under 2 ms and a maximum spike sorting rate of 941 spikes/second have been achieved with our hybrid hardware/software system. The low sorting latency and fast sorting rate allow future system developments of neural circuit modulation through analyzing neural activities in real-time.