Small Subclusters Research Articles

Modelling star cluster formation: Mergers

Abstract Star cluster formation in giant molecular clouds involves the local collapse of the cloud into small gas-rich subclusters, which can then subsequently collide and merge to build up the final star cluster(s). In this paper, we simulate collisions between these subclusters, using coupled smooth particle hydrodynamics for the gas and N-body dynamics for the stars. We are guided by previous radiation hydrodynamics simulations of molecular cloud collapse which provide the global properties of the colliding clusters, such as their stellar and gas masses, and their initial positions and velocities. The subclusters in the original simulation were treated as sink particles which immediately merged into a single entity after the collision. We show that the more detailed treatment provides a more complex picture. At collisional velocities above ≈ 10 km/s, the stellar components of the cluster do not form a monolithic cluster within 3 Myr, although the gas may do so. At lower velocities, the clusters do eventually merge but over timescales that may be longer than the time for a subsequent collision. The structure of the resultant cluster is not well-fit by any standard density distribution, and the clusters are not in equilibrium but continue to expand over our simulation time. We conclude that the simple sink particle treatment of subcluster mergers in large-scale giant molecular cloud simulations provides an upper limit on the final cluster properties.

Consensus reaching process in large-scale group decision making based on opinion leaders

With the development of science and technology, large-scale group decision making based on social network has become a crucial issue, and opinion leaders play an increasingly important role in the process of opinion dissemination and information transmission by virtue of their own influence, even affect the direction of decision-making. This paper proposes a framework of consensus reaching process for large-scale groups based on opinion leaders. In this framework, we first use the fast unfolding algorithm to divide the large-scale group into several small sub-clusters, and identify the opinion leaders of each sub-cluster. Then, we aggregate the opinions of the opinion leaders in each sub-cluster and construct the group decision matrix with the opinions of each sub-cluster as a new whole. A feedback mechanism based on opinion leaders is developed to help the opinion leaders who do not reach a consensus change their opinions to promote consensus. Finally, a numerical example about emergency shelter selection is given to illustrate the effectiveness of the proposed approach.

SVDD boundary and DPC clustering technique-based oversampling approach for handling imbalanced and overlapped data

Imbalanced datasets classification remains an important domain in machine learning. Conventional supervised learning algorithms tend to be biased towards the majority class when addressing imbalanced datasets, thus providing poor classification results on the minority class. The learning task would become more challenging when there are overlapping and within-imbalance issues in imbalanced datasets, which are often the case and have been proven to play more negative impact on the classification performance than between-class imbalance. In this paper, we propose a novel SVDD (Support Vector Data Description) boundary and DPC (Density Peaks Clustering) clustering technique-based oversampling approach (SVDDDPCO) for handling imbalanced and overlapped data. The proposed approach first utilizes support vector data description (SVDD) model with greater penalty constant for the minority class than the majority class to generate the class boundary, and then identifies those misclassified majority or few minority instances by the boundary as potential overlapped or noisy ones and eliminates them. To address the within-balance issues, density peaks clustering (DPC) is used to cluster the minority instances due to its advantage of accurately identifying sub-clusters with different sizes and densities, which facilitates simultaneously combating between-class and within-class imbalance issues caused by various reasons. The minority instances are assigned with different weights inversely proportional to their local densities and their distances to the boundary and then the size for each identified sub-cluster to be oversampled is adaptively determined according to its own size and the weights of its elements. Such strategy aims to generate more synthetic minority instances for borderline and sparser small sub-clusters which are usually informative to the later learning tasks. Finally, oversampling is performed within each sub-cluster with different sizes to not only counteract the within imbalance but also avoid the generation of any noisy or overlapped synthetic instance. Extensive comparison results on various datasets showed the proposed method achieves statistically significant improvements in classification performance relative to state-of-the-art ones.

CBF transcription factors involved in the cold response of Camellia japonica (Naidong).

CBFs belong to the ERF subfamily of the AP2 supergene family and often play an important role in the cold acclimation of temperate plants. However, the role of CBFs in Camellia japonica (Naidong), the only Camellia japonica population found in the temperate zones of China, remains unclear. It is very important to study the genetic composition of C. japonica (Naidong) to adapt to low temperature for Camellia species. Using full-length transcriptome data, we identified four CjCBF genes that respond to cold stress and analyzed their evolutionary relationships, domains, and expression patterns. The phylogeny of CBFs of 19 angiosperms divided the genes into three categories, and the four CjCBFs belong to a small subcluster. The strong response of CjCBF1 to cold treatment and its sustained high level of expression indicated that it plays an important role in the process of cold acclimation. A yeast two-hybrid assay revealed an interaction between CjCBF1, CjCBF2, and CjCBF5, and subcellular localization confirmed this finding. The expression of CjCBFs was tissue-specific: CBF1 was mainly expressed in leaves, and CBF3 was mainly expressed in stem. The responses of the four CjCBFs to drought and high temperature and the effect of light were also characterized. Our study provides new insight into the role of CBFs in the cold response in C. japonica (Naidong).

Divide well to merge better: A novel clustering algorithm

In this paper, a novel non-parametric clustering algorithm which is based on the concept of divide-and-merge is proposed. The proposed algorithm is based on two primary phases, after data cleaning: (i) the Division phase and (ii) the Merging phase. In the initial phase of division, the data is divided into an optimized number of small sub-clusters utilizing all the dimensions of the data. In the second phase of merging, the small sub-clusters obtained as a result of division are merged according to an advanced statistical metric to form the actual clusters in the data. The proposed algorithm has the following merits: (i) ability to discover both convex and non-convex shaped clusters, (ii) ability to discover clusters different in densities, (iii) ability to detect and remove outliers/noise in the data (iv) easily tunable or fixed hyperparameters (v) and its usability for high dimensional data. The proposed algorithm is extensively tested on 20 benchmark datasets including both, the synthetic and the real datasets and is found better/competing to the existing state-of-the-art parametric and non-parametric clustering algorithms.

Mobility degradation in 4H-SiC MOSFETs and interfacial formation of carbon clusters

We describe a detailed atomic model of the SiC/SiO2 interface states due to carbon clusters which lower the field effect mobility of SiC below its Hall effect value, due to the oxidation process. The carbon clusters are on the SiO2 side. We show for the first time that carbon clusters containing a mixture of sp2 and sp3 sites explain the results, with these sp2 sites forming small sub-clusters to maintain a large local band gap.

The Variability of Qualitative Traits in Promising Rice Lines

Rice is one of the world’s oldest important crops. Qualitative traits can be useful for the identification of promising rice lines. In this study, seventeen advance rice lines were evaluated for qualitative traits. This assessment was carried out to analyse the variability in genotypes, based on 49 important qualitative traits of rice. Data on morphological parameters was collected from each line at suitable phase of growth to check variation. Findings were recorded according to descriptors for rice. Dendrogram was constructed which divided all the genotypes into seventeen major clusters, each subdivided into smaller sub-clusters. The pattern of cluster based on qualitative traits revealed the wide genetic variation. A high level of genetic variability based on qualitative traits was observed in this study. The genetic improvement of the rice needs the availability of genetic variability in promising lines. Significant phenotypic information and these variability has potential to develop quality rice for future breeding programs.

A multi-stage hierarchical clustering algorithm based on centroid of tree and cut edge constraint

The minimum spanning tree clustering algorithm is known to be capable of detecting clusters with irregular boundaries. The paper presents a novel hierarchical clustering algorithm based on minimum spanning tree (MST), which tends to reduce the complexity of the merging process with guaranteed clustering performance. There are two core ideas in the proposed method: (1) The inter-cluster distance is calculated with the centroid of MST instead of the center of cluster. (2) The length of cut edge at the intersection of two adjacent clusters is taken as a merge condition. Based on this idea, we propose a three-stage MST-based hierarchical clustering algorithm (CTCEHC). In Stage 1, a preliminary partition is performed with the degrees of vertices in MST. In Stage 2, small subclusters are merged via the geodesic distance between the centroids of MST in two clusters and the cut edge constraint I. In Stage 3, the adjacent cluster pairs satisfying the cut edge constraint II are merged. The experimental results on the synthetic data sets and real data sets demonstrate a good performance of the proposed clustering method.

A Spatial Ecological Study Based on Benthic Foraminifera off Central West Coast of India: An Approach through Cluster Analysis with Special Reference to Oxygen Minimum Zone

Fifty-two surface samples collected from a depth range between 15 and 3300 m from the Vijaydurg-Karwar sector off the central west coast of India were analyzed for benthic foraminifera. A total of 386 benthic foraminifera were identified to species level. The relative abundances of 21 major genera (>5% in at least two stations) were plotted on a distribution map as well as on a kite diagram. Most of the genera show correlation to water depth in the study area. For better understanding and to draw meaningful conclusions from the assemblages out of the surface distribution of major genera, Q-mode cluster analysis was performed with the total number of species as variables. The clusters (namely A & B) and their subclusters (namely B1, B2 & B3) also exhibit a preference to water depth. Considering the position of the Oxygen Minimum Zone (OMZ) in this region extends from 150 to 1500 m water depth, it was noticed that the stations in smaller subcluster within cluster B3 fall below the OMZ, while most of the stations in cluster B2 fall above the OMZ. From the cluster analysis, the major benthic foraminiferal species from within the OMZ are identified as Bolivina silvestrina, Bulimina marginata, Globocassidulina oriangulata and Melonis pompilioides. Hence, cluster analysis based on the benthic foraminiferal assemblage is useful to detect the position of the OMZ in the study area.

New imbalanced fault diagnosis framework based on Cluster-MWMOTE and MFO-optimized LS-SVM using limited and complex bearing data

Due to the complexity of their working conditions, historical rolling bearing datasets are mostly limited and imbalanced. The fault data may be composed of multiple subclusters; that is, the historical rolling bearing data have both between-class and within-class imbalances. While support vector machines (e.g., least squares support vector machines (LS-SVMs)) offer advantages when dealing with limited data, traditional fault diagnosis using an LS-SVM has the disadvantages of easy failure of complex imbalanced data and large dependence on the classifier hyperparameters. Therefore, this paper presents a new imbalanced fault diagnosis framework based on a cluster-majority weighted minority oversampling technique (Cluster-MWMOTE) and a moth-flame optimization (MFO)-based LS-SVM classifier. As an extension of MWMOTE, our proposed Cluster-MWMOTE combines the clustering algorithm represented by agglomerative hierarchical clustering (AHC) with MWMOTE. Unlike MWMOTE, Cluster-MWMOTE can avoid the ignoring of small subclusters of faulty (minority) instances far from normal (majority) instances. That is, Cluster-MWMOTE further improves the adaptation to within-class imbalances. As a novel heuristic intelligent algorithm, MFO exhibits faster convergence and higher precision than the traditional optimization algorithms (e.g., particle swarm optimization (PSO) and genetic algorithm (GA)). Therefore, we utilize MFO to optimize the hyperparameters (Sigma &γ) of the LS-SVM classifier for the first time. The fault diagnosis results represented by CWRU and IMS bearing data suggest that the proposed framework provides higher fault diagnosis recognition rates and algorithm robustness than 16 existing algorithms.

Clustered nuclei splitting based on recurrent distance transform in digital pathology images

The accuracy of the applied technique for automated nuclei segmentation is critical in obtaining high-quality and efficient diagnostic results. Unfortunately, multiple objects in histopathological images are connected (clustered) and frequently counted as one. In this study, we present a new method for cluster splitting based on distance transform binarized with the recurrently increased threshold value and modified watershed algorithm. The proposed method treats clusters separately, splitting them into smaller sub-clusters and conclusively into separate objects, based solely on the shape feature, making it independent of the pixel intensity. The efficiency of these algorithms is validated based on the labeled set of images from two datasets: BBBC004v1 and breast cancer tissue microarrays. Results of initial nuclei detection were significantly improved by applying the proposed algorithms. Our approach outperformed the state-of-the-art techniques based on recall, precision, F1-score, and Jaccard index. The proposed method achieves very low amount of under-segmented, as well as over-segmented objects. In summary, we provide novel and efficient method for dividing the clustered nuclei in digital images of histopathological slides.

Structural and electronic properties of nanosize semiconductor CeSin0/− (n = 4–20) material: A double-hybrid density functional theory investigation

Structural evolution model, HOMO-LUMO gap, relative stability, magnetic moment and spectral property of nanosize CeSin0/− (n = 4–20) semiconductor material were performed by employing double-hybrid density functional scheme. Optimized geometries for CeSin displayed that the most stable structure is Ce-substituted for a Si in the ground state of Sin+1 with n ≤ 17; Ce atom as linker connects two small silicon subclusters with n = 17–19; and Ce atom is encapsulated into cage-like Si framework with ≥20. For anion, the most stable structure of CeSin− is Ce-linked configuration with n = 8–18 excluded n = 9, and Ce atom is encapsulated into cage-like motif with n = 19–20. The spectral information was reported. The 4f electron of Ce atom in CeSin0/− does not participate in bonding, resulting the total magnetic moments of neutral and anionic CeSin0/− clusters are 2 and 1 μB, respectively. HOMO-LUMO gap, relative stability and chemical bonding analyses revealed that encapsulated configuration of CeSi20− is a “magic” cluster.

Sloshing of Galaxy Cluster Core Plasma in the Presence of Self-interacting Dark Matter

Abstract The “sloshing” of the cold gas in the cores of relaxed clusters of galaxies is a widespread phenomenon evidenced by the presence of spiral-shaped “cold fronts” in X-ray observations of these systems. In simulations, these flows of cold gas readily form via interactions of the cluster core with small subclusters, due to a separation of the cold gas from the dark matter (DM), due to their markedly different collisionalities. In this work, we use numerical simulations to investigate the effects of increasing the DM collisionality on sloshing cold fronts in a cool-core cluster. For clusters in isolation, the formation of a flat DM core via self-interactions results in modest adiabatic expansion and cooling of the core gas. In merger simulations, cold fronts form in the same manner as in previous simulations, but the flattened potential in the core region enables the gas to expand to larger radii in the initial stages. Upon infall, the subclusters DM mass decreases via collisions, reducing its influence on the core. Thus, the sloshing gas moves slower, inhibiting the growth of fluid instabilities relative to simulations where the DM cross section is zero. This also inhibits turbulent mixing and the increase in entropy that would otherwise result. For values of the cross section σ/m ≳ 1, subclusters do not survive as self-gravitating structures for more than two core passages. Additionally, separations between the peaks in the X-ray emissivity and thermal Sunyaev–Zeldovich effect signals during sloshing may place constraints on DM self-interactions.

Mapping the chemical and sequence space of the ShKT superfamily

The ShKT superfamily is widely distributed throughout nature and encompasses a wide range of documented functions and processes, from modulation of potassium channels to involvement in morphogenesis pathways. Cysteine-rich secretory proteins (CRISPs) contain a cysteine-rich domain (CRD) at the C-terminus that is similar in structure to the ShK fold. Despite the structural similarity of the CRD and ShK-like domains, we know little of the sequence-function relationships in these families. Here, for the first time, we examine the evolution of the biophysical properties of sequences within the ShKT superfamily in relation to function, with a focus on the ShK-like superfamily. ShKT data were sourced from published sequences in the protein family database, in addition to new ShK-like sequences from the Australian speckled anemone (Oulactis sp.). Our analysis clearly delineates the ShK-like family from the CRDs of CRISP proteins. The four CRISP subclusters separate out into the main phyla of Mammalia, Insecta and Reptilia. The ShK-like family is in turn composed of seven subclusters, the largest of which contains members from across the eukaryotes, with a continuum of intermediate properties. Smaller sub-clusters contain specialised members such as nematode ShK-like sequences. Several of these ShKT sub-clusters contain no functionally characterised sequences. This chemical space analysis should be useful as a guide to select sequences for functional studies and to gain insight into the evolution of these highly divergent sequences with an ancient conserved fold.

2D melting of confined colloids with a mixture of square and triangular order

We implement Brownian dynamics to investigate the melting processes of colloidal particles confined isotropically and interacting via a potential which can be tailored in a repulsive–attractive–respulsive fashion as the interparticle distance increases. The stable configurations of such a system is composed of a large diversity of structures, which includes quasicrystalline, triangular, square, and mixed orderings, as well as the presence of fringes and holes, which are located, respectively, at the border and interior of the clusters. Our simulations demonstrate that during the melting process particles are able to swing between different micro phases. This intermediary stage, present in a finite range of temperature, precedes the melting in all cases investigated and is different from the hexatic phase of the KTNHY framework. We also test the fringes stability and find it to be higher than the one found in compact clusters. Finally, we show that, at the high temperature regime, the system loses its angular ordering while still preserves its radial interparticle confinement, which, ultimately, causes the proliferation of small subclusters.

The Structure of the Young Star Cluster NGC 6231. II. Structure, Formation, and Fate

Abstract The young cluster NGC 6231 (stellar ages ∼2–7 Myr) is observed shortly after star formation activity has ceased. Using the catalog of 2148 probable cluster members obtained from Chandra, VVV, and optical surveys (Paper I), we examine the cluster’s spatial structure and dynamical state. The spatial distribution of stars is remarkably well fit by an isothermal sphere with moderate elongation, while other commonly used models like Plummer spheres, multivariate normal distributions, or power-law models are poor fits. The cluster has a core radius of 1.2 ± 0.1 pc and a central density of ∼200 stars pc−3. The distribution of stars is mildly mass segregated. However, there is no radial stratification of the stars by age. Although most of the stars belong to a single cluster, a small subcluster of stars is found superimposed on the main cluster, and there are clumpy non-isotropic distributions of stars outside ∼4 core radii. When the size, mass, and age of NGC 6231 are compared to other young star clusters and subclusters in nearby active star-forming regions, it lies at the high-mass end of the distribution but along the same trend line. This could result from similar formation processes, possibly hierarchical cluster assembly. We argue that NGC 6231 has expanded from its initial size but that it remains gravitationally bound.

Characterization and identification of SSR markers for screening of cowpea genotypes against Bean common mosaic virus (BCMV) disease resistance

One hundred cowpea genotypes were screened against Bean common mosaic virus (BCMV) by artificial inoculation genotypes. The genotypes showed different disease reaction viz., immune, resistant, moderately resistant, Sap moderately susceptible, susceptible and highly susceptible. The SSR markers were characterized using a total of 25 randomly selected resistance and susceptible cowpea genotypes. Among, 10 SSR markers used to screen polymorphism, only 4 primers were found to be polymorphic and they were used for characterization and finding out genetic relationship of cowpea genotypes. The polymorphism information content (PIC) of these SSR markers ranged from 0.23 to 0.37 and single marker analysis (SMA) revealed, out of these 4 SSR markers M 135 marker explained 28.2 % of phenotypic variation (R2). The banding pattern of all cowpea genotypes were scored and dendrogram was constructed. A dendrogram of these genotypes based on SSR polymorphism divided into two major clusters at 35 % similarity. Cluster I was the biggest with 20 genotypes which was further separated as three sub-clusters. Whereas, 5 genotypes grouped in cluster II with two small sub-clusters. The clustering pattern of resistant and susceptible genotypes was not clear, however resistant genotypes such as V-5 and IC 8996 clustered in one group with 100 % similarity. The information provided by the identified markers would be very useful in breeding programs to select cowpea lines resistance to BCMV disease and M 135 was found to be good for rapid and accurate screening of cowpea genotypes against BCMV disease resistance.

A Novel clustering method based on hybrid K-nearest-neighbor graph

Most of the existing clustering methods have difficulty in processing complex nonlinear data sets. To remedy this deficiency, in this paper, a novel data model termed Hybrid K-Nearest-Neighbor (HKNN) graph, which combines the advantages of mutual k-nearest-neighbor graph and k-nearest-neighbor graph, is proposed to represent the nonlinear data sets. Moreover, a Clustering method based on the HKNN graph (CHKNN) is proposed. The CHKNN first generates several tight and small subclusters, then merges these subclusters by exploiting the connectivity among them. In order to select the optimal parameters for CHKNN, we further propose an internal validity index termed K-Nearest-Neighbor Index (KNNI), which can also be used to evaluate the validity of nonlinear clustering results by varying a control parameter. Experimental results on synthetic and real-world data sets, as well as that on the video clustering, have demonstrated the significant improvement on performance over existing nonlinear clustering methods and internal validity indices.

Star formation triggered by non-head-on cloud–cloud collisions, and clouds with pre-collision sub-structure

In an earlier paper, we used smoothed particle hydrodynamics (SPH) simulations to explore star formation triggered by head-on collisions between uniform-density 500M(circle dot) clouds, and showed that there is a critical collision velocity, v(CRIT). At collision velocities below vCRIT, a hub-and-spoke mode operates and delivers a monolithic cluster with a broad mass function, including massive stars (M-star greater than or similar to 10M(circle dot)) formed by competitive accretion. At collision velocities above vCRIT, a spider's-web mode operates and delivers a loose distribution of small sub-clusters with a relatively narrow mass function and no massive stars. Here we show that, if the head-on assumption is relaxed, vCRIT is reduced. However, if the uniform-density assumption is also relaxed, the collision velocity becomes somewhat less critical: a low collision velocity is still needed to produce a global hub-and-spoke system and a monolithic cluster, but, even at high velocities, large cores -capable of supporting competitive accretion and thereby producing massive stars - can be produced. We conclude that cloud-cloud collisions may be a viable mechanism for forming massive stars - and we show that this might even be the major channel for forming massive stars in the Galaxy.

Spatial clustering with Density-Ordered tree

Abstract Clustering has emerged as an active research direction for knowledge discovery in spatial databases. Most spatial clustering methods become ineffective when inappropriate parameters are given or when datasets of diverse shapes and densities are provided. To address this issue, we propose a novel clustering method, called SCDOT (Spatial Clustering with Density-Ordered Tree). By projecting a dataset to a Density-Ordered Tree, SCDOT partitions the data into several relatively small sub-clusters with a box-plot method. A heuristic method is proposed to find the genuine clusters by repeatedly merging sub-clusters and an iteration strategy is utilized to automatically determine input parameters. Moreover, we also provide an innovative way to identify cluster center and noise. Extensive experiments on both synthetic and real-world datasets demonstrate the superior performance of SCDOT over the baseline methods.