Channel Clustering Research Articles

Engineering a K+ channel 'sensory antenna' enhances stomatal kinetics, water use efficiency and photosynthesis.

Stomata of plant leaves open to enable CO2 entry for photosynthesis and close to reduce water loss via transpiration. Compared with photosynthesis, stomata respond slowly to fluctuating light, reducing assimilation and water use efficiency. Efficiency gains are possible without a cost to photosynthesis if stomatal kinetics can be accelerated. Here we show that clustering of the GORK channel, which mediates K+ efflux for stomatal closure in the model plant Arabidopsis, arises from binding between the channel voltage sensors, creating an extended 'sensory antenna' for channel gating. Mutants altered in clustering affect channel gating to facilitate K+ flux, accelerate stomatal movements and reduce water use without a loss in biomass. Our findings identify the mechanism coupling channel clustering with gating, and they demonstrate the potential for engineering of ion channels native to the guard cell to enhance stomatal kinetics and improve water use efficiency without a cost in carbon fixation.

Three-dimensional visualization of the cardiac ryanodine receptor clusters and the molecular-scale fraying of dyads.

Clusters of ryanodine receptor calcium channels (RyRs) form the primary molecular machinery of intracellular calcium signalling in cardiomyocytes. While a range of optical super-resolution microscopy techniques have revealed the nanoscale structure of these clusters, the three-dimensional (3D) nanoscale topologies of the clusters have remained mostly unresolved. In this paper, we demonstrate the exploitation of molecular-scale resolution in enhanced expansion microscopy (EExM) along with various 2D and 3D visualization strategies to observe the topological complexities, geometries and molecular sub-domains within the RyR clusters. Notably, we observed sub-domains containing RyR-binding protein junctophilin-2 (JPH2) occupying the central regions of RyR clusters in the deeper interior of the myocytes (including dyads), while the poles were typically devoid of JPH2, lending to a looser RyR arrangement. By contrast, peripheral RyR clusters exhibited variable co-clustering patterns and ratios between RyR and JPH2. EExM images of dyadic RyR clusters in right ventricular (RV) myocytes isolated from rats with monocrotaline-induced RV failure revealed hallmarks of RyR cluster fragmentation accompanied by breaches in the JPH2 sub-domains. Frayed RyR patterns observed adjacent to these constitute new evidence that the destabilization of the RyR arrays inside the JPH2 sub-domains may seed the primordial foci of dyad remodelling observed in heart failure.This article is part of the theme issue ‘The cardiomyocyte: new revelations on the interplay between architecture and function in growth, health, and disease’.

Real-Time Fast Channel Clustering for LiDAR Point Cloud

LiDAR sensors can produce point clouds with precise 3D depth information that is essential for autonomous vehicles and robotic systems. As a perception task, point cloud clustering algorithms can be applied to segment the points into object instances. In this brief, we propose a novel, hardware-friendly fast channel clustering (FCC) algorithm that achieves state-of-the-art performance when evaluated using KITTI panoptic segmentation benchmark. Furthermore, an efficient, pipeline hardware architecture is proposed to implement the FCC algorithm on an FPGA. Experiments show that the hardware design can process each LiDAR frame with 64 channels, 2048 horizontal resolution at various point sparsity in 1.93 ms, which is more than 471.5 times faster than running on the CPU. The code will be released to the public via GitHub.

Adaptable and Robust EEG Bad Channel Detection Using Local Outlier Factor (LOF).

Electroencephalogram (EEG) data are typically affected by artifacts. The detection and removal of bad channels (i.e., with poor signal-to-noise ratio) is a crucial initial step. EEG data acquired from different populations require different cleaning strategies due to the inherent differences in the data quality, the artifacts’ nature, and the employed experimental paradigm. To deal with such differences, we propose a robust EEG bad channel detection method based on the Local Outlier Factor (LOF) algorithm. Unlike most existing bad channel detection algorithms that look for the global distribution of channels, LOF identifies bad channels relative to the local cluster of channels, which makes it adaptable to any kind of EEG. To test the performance and versatility of the proposed algorithm, we validated it on EEG acquired from three populations (newborns, infants, and adults) and using two experimental paradigms (event-related and frequency-tagging). We found that LOF can be applied to all kinds of EEG data after calibrating its main hyperparameter: the LOF threshold. We benchmarked the performance of our approach with the existing state-of-the-art (SoA) bad channel detection methods. We found that LOF outperforms all of them by improving the F1 Score, our chosen performance metric, by about 40% for newborns and infants and 87.5% for adults.

Active forgetting requires Sickie function in a dedicated dopamine circuit in Drosophila

Forgetting is an essential component of the brain's memory management system, providing a balance to memory formation processes by removing unused or unwanted memories, or by suppressing their expression. However, the molecular, cellular, and circuit mechanisms underlying forgetting are poorly understood. Here we show that the memory suppressor gene, sickie, functions in a single dopamine neuron (DAn) by supporting the process of active forgetting in Drosophila. RNAi knockdown (KD) of sickie impairs forgetting by reducing the Ca2+ influx and DA release from the DAn that promotes forgetting. Coimmunoprecipitation/mass spectrometry analyses identified cytoskeletal and presynaptic active zone (AZ) proteins as candidates that physically interact with Sickie, and a focused RNAi screen of the candidates showed that Bruchpilot (Brp)-a presynaptic AZ protein that regulates calcium channel clustering and neurotransmitter release-impairs active forgetting like sickie KD. In addition, overexpression of brp rescued the impaired forgetting of sickie KD, providing evidence that they function in the same process. Moreover, we show that sickie KD in the DAn reduces the abundance and size of AZ markers but increases their number, suggesting that Sickie controls DAn activity for forgetting by modulating the presynaptic AZ structure. Our results identify a molecular and circuit mechanism for normal levels of active forgetting and reveal a surprising role of Sickie in maintaining presynaptic AZ structure for neurotransmitter release.

Increased KV2.1 Channel Clustering Underlies the Reduction of Delayed Rectifier K+ Currents in Hippocampal Neurons of the Tg2576 Alzheimer's Disease Mouse.

Alzheimer’s disease (AD) is a neurodegenerative disorder characterized by the progressive deterioration of cognitive functions. Cortical and hippocampal hyperexcitability intervenes in the pathological derangement of brain activity leading to cognitive decline. As key regulators of neuronal excitability, the voltage-gated K+ channels (KV) might play a crucial role in the AD pathophysiology. Among them, the KV2.1 channel, the main α subunit mediating the delayed rectifier K+ currents (IDR) and controlling the intrinsic excitability of pyramidal neurons, has been poorly examined in AD. In the present study, we investigated the KV2.1 protein expression and activity in hippocampal neurons from the Tg2576 mouse, a widely used transgenic model of AD. To this aim we performed whole-cell patch-clamp recordings, Western blotting, and immunofluorescence analyses. Our Western blotting results reveal that KV2.1 was overexpressed in the hippocampus of 3-month-old Tg2576 mice and in primary hippocampal neurons from Tg2576 mouse embryos compared with the WT counterparts. Electrophysiological experiments unveiled that the whole IDR were reduced in the Tg2576 primary neurons compared with the WT neurons, and that this reduction was due to the loss of the KV2.1 current component. Moreover, we found that the reduction of the KV2.1-mediated currents was due to increased channel clustering, and that glutamate, a stimulus inducing KV2.1 declustering, was able to restore the IDR to levels comparable to those of the WT neurons. These findings add new information about the dysregulation of ionic homeostasis in the Tg2576 AD mouse model and identify KV2.1 as a possible player in the AD-related alterations of neuronal excitability.

Complexity analysis of quantum teleportation via different entangled channels in the presence of noise

Quantum communication is one of the hot topics in quantum computing, where teleportation of a quantum state has a slight edge and gained significant attention from researchers. A large number of teleportation schemes have already been introduced so far. Here, we compare the teleportation of a single qubit message among different entangled channels such as the two-qubit Bell channel, three-qubit GHZ channel, two- and three-qubit cluster states, the highly entangled five-qubit Brown \emph{et al.} state and the six-qubit Borras \emph{et al.} state. We calculate and compare the quantum costs in each of the cases. Furthermore, we study the effects of six noise models, namely bit-flip noise, phase-flip noise, bit-phase flip noise, amplitude damping, phase damping and the depolarizing error that may affect the communication channel used for the teleportation. An investigation on the variation of the initial state's fidelity with respect to the teleported state in the presence of the noise model is performed. A visual representation of the variation of fidelity for various values of the noise parameter $\eta$ is done through a graph plot. It is observed that as the value of noise parameter in the range $\eta \in [0,0.5]$, the fidelity decreases in all the entangled channels under all the noise models. After that, in the Bell channel, GHZ channel and three-qubit cluster state channel, the fidelity shows an upward trend under all the noise models. However, in the other three channels, the fidelity substantially decreases in the case of amplitude damping, phase damping and depolarizing noise, and even it reaches zero for $\eta = 1$ in Brown \emph{et al.} and Borras \emph{et al.} channels.

Crystal structure of Ankyrin-G in complex with a fragment of Neurofascin reveals binding mechanisms required for integrity of the axon initial segment

The axon initial segment (AIS) has characteristically dense clustering of voltage-gated sodium channels (Nav), cell adhesion molecule Neurofascin 186 (Nfasc), and neuronal scaffold protein Ankyrin-G (AnkG) in neurons, which facilitates generation of an action potential and maintenance of axonal polarity. However, the mechanisms underlying AIS assembly, maintenance, and plasticity remain poorly understood. Here, we report the high-resolution crystal structure of the AnkG ankyrin repeat (ANK repeat) domain in complex with its binding site in the Nfasc cytoplasmic tail that shows, in conjunction with binding affinity assays with serial truncation variants, the molecular basis of AnkG–Nfasc binding. We confirm AnkG interacts with the FIGQY motif in Nfasc, and we identify another region required for their high affinity binding. Our structural analysis revealed that ANK repeats form 4 hydrophobic or hydrophilic layers in the AnkG inner groove that coordinate interactions with essential Nfasc residues, including F1202, E1204, and Y1212. Moreover, we show disruption of the AnkG–Nfasc complex abolishes Nfasc enrichment at the AIS in cultured mouse hippocampal neurons. Finally, our structural and biochemical analysis indicated that L1 syndrome-associated mutations in L1CAM, a member of the L1 immunoglobulin family proteins including Nfasc, L1CAM, NrCAM, and CHL1, compromise binding with ankyrins. Taken together, these results define the mechanisms underlying AnkG–Nfasc complex formation and show that AnkG-dependent clustering of Nfasc is required for AIS integrity.

Muscle Synergy-driven Motor Unit Clustering for Human-Machine Interfacing.

Electromyographic signals (EMGs) can provide information on the overall activity of the innervating motor neuros in any given muscle but also globally reflect the underlying neuromechanics of human movement (e.g., muscle synergies). motor unit(MU) decomposition is a technique based on the deconvolution of high-density EMGs (HD-EMG) in order to derive the activities of the corresponding motor neurons. This powerful yet very sensitive tool has seen some traction in human-machine interfacing (HMI) for rehabilitation. Here, we propose combining the synergy-inspired channel clustering in order to isolate the most prominent regions of EMG activation in each targeted degree of freedom (DoF) and thus cater to decomposition's sensitivity demands. Our assumption is that this will lead to a higher number of extracted MUs and consequently better motion estimation in HMIs. Indeed, in four subjects, we have shown a 69% average increase in the number of MUs when decomposition was done using muscle-synergy channel clustering. Consequently, all three of our kinematic estimators benefited from an increased pool of units, with the linear regressor showing the greatest improvement once compared to, the artificial neural network and the gated recurrent unit, which had the overall best performance. Clinical Relevance- The results demonstrated in this work provide a new perspective on the online EMG-driven HMI systems that can be greatly beneficial in the rehabilitation of motor disorders.

Studying KcsA Channel Clustering Using Single Channel Voltage-Clamp Fluorescence Imaging.

Oligomerization and complex formation play a key role for many membrane proteins and has been described to influence ion channel function in both neurons and the heart. In this study, we observed clustering of single KcsA channels in planar lipid bilayer using single molecule fluorescence, while simultaneously measuring single channel currents. Clustering coincided with cooperative opening of KcsA. We demonstrate that clustering was not caused by direct protein-protein interactions or hydrophobic mismatch with the lipid environment, as suggested earlier, but was mediated via microdomains induced by the channel in the lipid matrix. We found that single channel activity of KcsA requires conically-shaped lipids in the lamellar liquid-crystalline (Lα) phase, and the need for a negative spontaneous curvature seem to lead to the deformations in the membrane that cause the clustering. The method introduced here will be applicable to follow oligomerization of a wide range of membrane proteins.

Machine Learning-Based Multipath Components Clustering and Cluster Characteristics Analysis in High-Speed Railway Scenarios

Channel multipath components (MPCs) clustering and cluster characterization are the prerequisite for the development of cluster based channel models. This article investigates the MPCs clustering based on machine learning (ML) and analyzes the cluster characteristics in typical high-speed railway (HSR) scenarios. A variational Bayesian Gaussian mixture model (GMM)-based algorithm is introduced to achieve the space–time clustering of MPCs, which incorporates the statistical characteristics of MPCs and can automatically determine the optimum number of clusters. Moreover, a novel density-based validity index is proposed for evaluating the MPCs clustering performance. The proposed validity index improves the traditional index by considering the intracluster density, which can be calculated according to the Graham scanning method and Green’s formula. In addition to synthetic datasets, realistic MPCs datasets collected in an HSR obstructed viaduct scenario are used for the performance evaluation of the clustering algorithm and clustering validity index. Based on clustering results in the measured scenario, static cluster characteristics, including cluster number, intercluster and intracluster delay spreads (DSs) and angle spreads (ASs), and dynamic cluster characteristics such as cluster lifetime and birth and death property of clusters, are extracted and analyzed. These results will be useful for cluster-based channel modeling in future HSR mobile communication systems.

Multiscale modeling of presynaptic dynamics from molecular to mesoscale.

Chemical synapses exhibit a diverse array of internal mechanisms that affect the dynamics of transmission efficacy. Many of these processes, such as release of neurotransmitter and vesicle recycling, depend strongly on activity-dependent influx and accumulation of Ca2+. To model how each of these processes may affect the processing of information in neural circuits, and how their dysfunction may lead to disease states, requires a computationally efficient modelling framework, capable of generating accurate phenomenology without incurring a heavy computational cost per synapse. Constructing a phenomenologically realistic model requires the precise characterization of the timing and probability of neurotransmitter release. Difficulties arise in that functional forms of instantaneous release rate can be difficult to extract from noisy data without running many thousands of trials, and in biophysical synapses, facilitation of per-vesicle release probability is confounded by depletion. To overcome this, we obtained traces of free Ca2+ concentration in response to various action potential stimulus trains from a molecular MCell model of a hippocampal Schaffer collateral axon. Ca2+ sensors were placed at varying distance from a voltage-dependent calcium channel (VDCC) cluster, and Ca2+ was buffered by calbindin. Then, using the calcium traces to drive deterministic state vector models of synaptotagmin 1 and 7 (Syt-1/7), which respectively mediate synchronous and asynchronous release in excitatory hippocampal synapses, we obtained high-resolution profiles of instantaneous release rate, to which we applied functional fits. Synchronous vesicle release occurred predominantly within half a micron of the source of spike-evoked Ca2+ influx, while asynchronous release occurred more consistently at all distances. Both fast and slow mechanisms exhibited multi-exponential release rate curves, whose magnitudes decayed exponentially with distance from the Ca2+ source. Profile parameters facilitate on different time scales according to a single, general facilitation function. These functional descriptions lay the groundwork for efficient mesoscale modelling of vesicular release dynamics.

Biological noise is a key determinant of the reproducibility and adaptability of cardiac pacemaking and EC coupling.

Each heartbeat begins with the generation of an action potential in pacemaking cells in the sinoatrial node. This signal triggers contraction of cardiac muscle through a process termed excitation-contraction (EC) coupling. EC coupling is initiated in dyadic structures of cardiac myocytes, where ryanodine receptors in the junctional sarcoplasmic reticulum come into close apposition with clusters of CaV1.2 channels in invaginations of the sarcolemma. Cooperative activation of CaV1.2 channels within these clusters causes a local increase in intracellular Ca2+ that activates the juxtaposed ryanodine receptors. A salient feature of healthy cardiac function is the reliable and precise beat-to-beat pacemaking and amplitude of Ca2+ transients during EC coupling. In this review, we discuss recent discoveries suggesting that the exquisite reproducibility of this system emerges, paradoxically, from high variability at subcellular, cellular, and network levels. This variability is attributable to stochastic fluctuations in ion channel trafficking, clustering, and gating, as well as dyadic structure, which increase intracellular Ca2+ variance during EC coupling. Although the effects of these large, local fluctuations in function and organization are sometimes negligible at the macroscopic level owing to spatial-temporal summation within and across cells in the tissue, recent work suggests that the "noisiness" of these intracellular Ca2+ events may either enhance or counterintuitively reduce variability in a context-dependent manner. Indeed, these noisy events may represent distinct regulatory features in the tuning of cardiac contractility. Collectively, these observations support the importance of incorporating experimentally determined values of Ca2+ variance in all EC coupling models. The high reproducibility of cardiac contraction is a paradoxical outcome of high Ca2+ signaling variability at subcellular, cellular, and network levels caused by stochastic fluctuations in multiple processes in time and space. This underlying stochasticity, which counterintuitively manifests as reliable, consistent Ca2+ transients during EC coupling, also allows for rapid changes in cardiac rhythmicity and contractility in health and disease.

Loss of \u03b24-spectrin impairs Nav channel clustering at the heminode and temporal fidelity of presynaptic spikes in developing auditory brain

Beta-4 (β4)-spectrin, encoded by the gene Sptbn4, is a cytoskeleton protein found at nodes and the axon initial segments (AIS). Sptbn4 mutations are associated with myopathy, neuropathy, and auditory deficits in humans. Related to auditory dysfunction, however, the expression and roles of β4-spectrin at axon segments along the myelinated axon in the developing auditory brain are not well explored. We found during postnatal development, β4-spectrin is critical for voltage-gated sodium channel (Nav) clustering at the heminode along the nerve terminal, but not for the formation of nodal and AIS structures in the auditory brainstem. Presynaptic terminal recordings in Sptbn4geo mice, β4-spectrin null mice, showed an elevated threshold of action potential and increased failures during action potential train at high-frequency. Sptbn4geo mice exhibited a slower central conduction and showed no startle responses, but had normal cochlear function. Taken together, the lack of β4-spectrin impairs Nav clustering at the heminode along the nerve terminal and the temporal fidelity and reliability of presynaptic spikes, leading to central auditory processing deficits during postnatal development.

CAMP-Dependent Synaptic Plasticity at the Hippocampal Mossy Fiber Terminal.

Cyclic adenosine monophosphate (cAMP) is a crucial second messenger involved in both pre- and postsynaptic plasticity in many neuronal types across species. In the hippocampal mossy fiber (MF) synapse, cAMP mediates presynaptic long-term potentiation and depression. The main cAMP-dependent signaling pathway linked to MF synaptic plasticity acts via the activation of the protein kinase A (PKA) molecular cascade. Accordingly, various downstream putative synaptic PKA target proteins have been linked to cAMP-dependent MF synaptic plasticity, such as synapsin, rabphilin, synaptotagmin-12, RIM1a, tomosyn, and P/Q-type calcium channels. Regulating the expression of some of these proteins alters synaptic release probability and calcium channel clustering, resulting in short- and long-term changes to synaptic efficacy. However, despite decades of research, the exact molecular mechanisms by which cAMP and PKA exert their influences in MF terminals remain largely unknown. Here, we review current knowledge of different cAMP catalysts and potential downstream PKA-dependent molecular cascades, in addition to non-canonical cAMP-dependent but PKA-independent cascades, which might serve as alternative, compensatory or competing pathways to the canonical PKA cascade. Since several other central synapses share a similar form of presynaptic plasticity with the MF, a better description of the molecular mechanisms governing MF plasticity could be key to understanding the relationship between the transcriptional and computational levels across brain regions.

SleepSEEG: automatic sleep scoring using intracranial EEG recordings only

Objective. To perform automatic sleep scoring based only on intracranial electroencephalography (iEEG), without the need for scalp EEG), electrooculography (EOG) and electromyography (EMG), in order to study sleep, epilepsy, and their interaction. Approach. Data from 33 adult patients was used for development and training of the automatic scoring algorithm using both oscillatory and non-oscillatory spectral features. The first step consisted in unsupervised clustering of channels based on feature variability. For each cluster the classification was done in two steps, a multiclass tree followed by binary classification trees to distinguish the more challenging stage N1. The test data consisted in 11 patients, in whom the classification was done independently for each channel and then combined to get a single stage per epoch. Main results. An overall agreement of 78% was observed in the test set between the sleep scoring of the algorithm using iEEG alone and two human experts scoring based on scalp EEG, EOG and EMG. Balanced sensitivity and specificity were obtained for the different sleep stages. The performance was excellent for stages W, N2, and N3, and good for stage R, but with high variability across patients. The performance for the challenging stage N1 was poor, but at a similar level as for published algorithms based on scalp EEG. High confidence epochs in different stages (other than N1) can be identified with median per patient specificity >80%. Significance. The automatic algorithm can perform sleep scoring of long-term recordings of patients with intracranial electrodes undergoing presurgical evaluation in the absence of scalp EEG, EOG and EMG, which are normally required to define sleep stages but are difficult to use in the context of intracerebral studies. It also constitutes a valuable tool to generate hypotheses regarding local aspects of sleep, and will be significant for sleep evaluation in clinical epileptology and neuroscience research.

Telegram and Digital Methods

Telegram and Digital Methods

Shank promotes action potential repolarization by recruiting BK channels to calcium microdomains.

Mutations altering the scaffolding protein Shank are linked to several psychiatric disorders, and to synaptic and behavioral defects in mice. Among its many binding partners, Shank directly binds CaV1 voltage activated calcium channels. Here, we show that the Caenorhabditis elegans SHN-1/Shank promotes CaV1 coupling to calcium activated potassium channels. Mutations inactivating SHN-1, and those preventing SHN-1 binding to EGL-19/CaV1 all increase action potential durations in body muscles. Action potential repolarization is mediated by two classes of potassium channels: SHK-1/KCNA and SLO-1 and SLO-2 BK channels. BK channels are calcium-dependent, and their activation requires tight coupling to EGL-19/CaV1 channels. SHN-1's effects on AP duration are mediated by changes in BK channels. In shn-1 mutants, SLO-2 currents and channel clustering are significantly decreased in both body muscles and neurons. Finally, increased and decreased shn-1 gene copy number produce similar changes in AP width and SLO-2 current. Collectively, these results suggest that an important function of Shank is to promote microdomain coupling of BK with CaV1.

Sleep Power Topography in Children with Attention Deficit Hyperactivity Disorder (ADHD).

Objective: Recent years saw an increasing interest towards sleep microstructure abnormalities in attention-deficit/hyperactivity disorder (ADHD). However, the existing literature on sleep electroencephalographic (EEG) power in ADHD is still controversial, often based on single electrode recordings, and mainly focused on slow wave activity (SWA) during NREM sleep. This study aimed to systematically investigate sleep power topography in all traditional frequency bands, in all sleep stages and across sleep cycles using high-density EEG (HD-EEG). Method: Thirty drug-naïve children with ADHD (10.5 ± 2.1 years, 21 male) and 23 typically developing (TD) control participants (mean age: 10.2 ± 1.6 years, 13 male) were included in the current analysis. Signal power topography was computed in classical frequency bands during sleep, contrasted between groups and sleep cycles, and correlated with measures of ADHD severity, cognitive functioning and estimated total sleep time. Results: Compared to TD subjects, patients with ADHD consistently displayed a widespread increase in low-frequency activity (between 3 and 10 Hz) during NREM sleep, but not during REM sleep and wake before sleep onset. Such a difference involved a wide centro-posterior cluster of channels in the upper SWA range, in Theta, and low-Alpha. Between-group difference was maximal in sleep stage N3 in the first sleep cycle, and positively correlated with average total sleep time. Conclusions: These results support the concept that children with ADHD, compared to TD peers, have a higher sleep pressure and altered sleep homeostasis, which possibly interfere with (and delay) cortical maturation.

Angular Clustering of Millimeter-Wave Propagation Channels With Watershed Transformation

An angular clustering method based on image processing is proposed in this paper. It is used to identify clusters in 2D representations of propagation channels. The approach uses operations such as watershed segmentation and is particularly well suited for clustering directional channels obtained by beam-steering at millimeter-wave. This situation occurs for instance with electronic beam-steering using analog antenna arrays during beam training process or during channel modeling measurements using either electronic or mechanical beam-steering. In particular, the proposed technique is used here to cluster two-dimensional power angular spectrum maps. The proposed clustering is unsupervised and is well suited to preserve the shape of clusters by considering the angular connection between neighbor samples, which is useful to obtain more accurate descriptions of channel angular properties. The approach is found to outperform approaches based on K-Power-Means in terms of accuracy as well as computational resources. The technique is assessed in simulation using IEEE 802.11ad channel model and in measurement using experiments conducted at 60 GHz in an indoor environment.