Distance-based Clustering Research Articles

Biosynthesis and Chemopreventive Potential of Jute (Corchorus capsularis and C. olitorius) Flavonoids and Phylogeny of Flavonoid Biosynthesis Pathways

AbstractFlavonoids are valuable phytochemicals for human health and nutrition. Jute (Corchorus capsularis and C. olitorius), a vegetable rich in phenolics and flavonoids, is globally consumed for its health benefit, but the biosynthesis pathways and metabolic profiles of its flavonoids are poorly characterized. Elucidating the flavonoid biosynthesis pathways would augment the broader use of jute, including targeted synthesis of its specific flavonoids. We reconstructed the core flavonoid biosynthesis pathways in jute by integrating transcriptome mining, HPLC and flavonoid histochemistry. In C. capsularis (white jute), the flavonoid biosynthesis pathways’ metabolic flux was driven toward the biosynthesis of proanthocyanidins that mediate the acquisition of abiotic stress tolerance. However, higher levels of flavonols in C. olitorius (tossa jute) render it more suitable for nutritional and medicinal use. Jute flavonoid extract exhibited in vitro inhibition of matrix metalloproteinase-2, suggesting its potential chemopreventive and immunity-boosting roles. Using the flavonoid biosynthesis pathways profiles of 93 plant species, we reconstructed the flavonoid biosynthesis pathways phylogeny based on distance-based clustering of reaction paths. This reaction-path flavonoid biosynthesis pathways phylogeny was quite distinct from that reconstructed using individual gene sequences. Our flavonoid biosynthesis pathways-based classification of flavonoid groups corroborates well with their chemical evolution, suggesting complex, adaptive evolution of flavonoid biosynthesis pathways, particularly in higher plants.

Genetic Variability and G×E Interactions in a Diverse Set of Groundnut Accessions

Background: Knowledge of the genetic diversity for various agronomic traits and their interaction with the environment and subsequent classification of genotypes will be beneficial for identification of divergent and stable sources of agronomic traits. Methods: A set of 96 groundnut germplasm accessions belonging to four botanical groups were evaluated for three years (2017 to 2019) for pod yield and component traits using AMMI analysis and subsequently accessions were classified based Euclidean cluster analysis. Result: Among different botanical groups, Virginia genotypes matured late and possessed high SPAD chlorophyll meter readings (SCMR) and pod yield compared to Spanish types. The component traits of pod maturity like days to flowering (first and 50%) showed low heritability and high genotype × environment interaction (GEI) and significant negatively affected sound mature kernel (SMK) and shelling per centage (SP). The cumulative contribution of environment and GEI component to the total variance was the highest in the expression of SP (67%) followed by days to maturity (54%) and days to 50% flowering (52%). Euclidean distance-based cluster analysis grouped the 96 accessions into five major clusters. Cluster I had accessions with higher pod yield, whereas cluster V contained accessions with low SLA, high SCMR and moderate pod yield. High yielding as well as stable accessions identified based on AMMI stability value (ASV) are NRCG 17332, 10076, 17268, 17197, 17108, 10106, 10089 and 17165. Trait specific as well as stable accessions identified in the present study can be useful donors for groundnut breeding programme.

MEDNet: Multiexpert Detection Network With Unsupervised Clustering of Training Samples

For various remote sensing objects, the current detection framework based on a single detection pipeline fails to provide satisfactory detection accuracy. In order to further improve the object recognition ability of the detection model, this article introduces the effective “multiexpert” mechanism into the field of remote sensing object detection and then constructs a multiexpert detection network (MEDNet). In this model, we first construct multiple feature pyramids (MFPs) to replace the traditional single feature pyramid to enrich the semantic representation ability of the model. Then, we equip multiple detection experts (MDEs) to leverage multiple kinds of features from MFP to perform different semantic predictions. As the first CNN-based multiexpert detection model for remote sensing images, we tailor a loss distance-based k-experts clustering (LD-kEC) strategy to assign training samples to different detection experts in an unsupervised fashion. By this strategy, we can directly use the existing remote sensing dataset without expert labels for end-to-end training of our multiexpert model. The experimental results prove that the proposed multiexpert-based detector can indeed significantly improve the object detection performance for remote sensing images.

Organization of big metrology data within the Cyber-Physical Manufacturing Metrology Model (CPM3)

In this paper, we propose a novel data curation concept that enables data mining and analytics within the recently described Cyber-Physical Manufacturing Metrology Model (CPM 3 ). The newly proposed methodology is based on organizing the metrology data into tree-based database structures using distance-based unsupervised clustering of the raw metrology data. Compared to traditionally utilized temporally organized lists, the new tree-based database organization of metrology data enables logarithmic acceleration of searches within the data and thus provides dramatic advantages for data mining. The newly proposed data curation methodology was evaluated in case studies involving hyper-spectral metrology of nanopatterned surfaces, coordinate measurement machine (CMM) inspection of aircraft engine turbines and imaging-based metrology of nano-volume droplets in the jet and fill stage of imprint lithography processes. Significant improvements in search speeds with minimal or no losses in search precision and recall were observed in all case-studies, with benefits of tree-based data organization growing with the size of the data.

An Innovative Perceptual Pigeon Galvanized Optimization (PPGO) Based Likelihood Naïve Bayes (LNB) Classification Approach for Network Intrusion Detection System

ABSTRACTIntrusion detection and classification have gained significant attention recently due to the increased utilization of networks. For this purpose, there are different types of Network Intrusion Detection System (NIDS) approaches developed in the conventional works, which mainly focus on identifying the intrusions from the datasets with the help of classification techniques. Still, it is limited by the significant problems of inefficiency in handling large dimensional datasets, high computational complexity, false detection, and more time consumption for training the models. To solve these problems, this research intends to develop an innovative clustering-based classification methodology to precisely detect intrusions from the different types of IDS datasets. Here, the most recent and extensively used IDS datasets such as NSL-KDD, CICIDS, and Bot-IoT have been employed for detecting intrusions. Data preprocessing has been performed to normalize the dataset to eliminate irrelevant attributes and organize the features. Then, the data separation is applied by forming the clusters by using an intelligent Anticipated Distance-based Clustering (ADC) incorporated with the Density-Based Spatial clustering of applications with noise (DBScan) algorithm. It helps to find the distance and density measures for grouping the attributes into the clusters, which increases the efficiency of classification. Here, the most suitable optimal parameters are selected using the Perpetual Pigeon Galvanized Optimization (PPGO) technique. The extracted features are used for training and testing the dataset samples.Consequently, the Likelihood Naove Bayes (LNB) classification approach is implemented to accurately predict the classified label as to whether normal or attack. During the evaluation, the performance of the proposed IDS framework is validated and compared using various evaluation metrics. Theresults show that the proposed ADC-DBScan-LNB model outperforms the other techniques with improved performance outcomes.

An Optimized K-means with Density and Distance-Based Clustering Algorithm for Multidimensional Spatial Databases

From massive and complex spatial database, the useful information and knowledge are extracted using spatial data mining. To analyze the complexity, efficient clustering algorithm for spatial database has been used in this area of research. The geographic areas containing spatial points are discovered using clustering methods in many applications. With spatial attributes, the spatial clustering problem have been designed using many approaches, but non-overlapping constraints are not considered. Most existing data mining algorithms suffer in high dimensions. With non-overlapping named as Non Overlapping Constraint based Optimized K-Means with Density and Distance-based Clustering (NOC-OKMDDC),a multidimensional optimization clustering is designed to solve this problem by the proposed system and the clusters with diverse shapes and densities in spatial databases are fast found. Proposed method consists of three main phases. Using weighted convolutional Neural Networks(Weighted CNN), attributes are reduced from the multidimensional dataset in this first phase. A partition-based algorithm (K-means) used by Optimized K-Means with Density and Distance-based Clustering (OKMDD) and several relatively small spherical or ball-shaped sub clusters are made by Clustering the dataset in this second phase. The optimal sub cluster count is performed with the help of Adaptive Adjustment Factor based Glowworm Swarm Optimization algorithm (AAFGSO). Then the proposed system designed an Enhanced Penalized Spatial Distance (EPSD) Measure to satisfy the non-overlapping condition. According to the spatial attribute values, the spatial distance between two points are well adjusted to achieving the EPSD. In third phase, to merge sub clusters the proposed system utilizes the Density based clustering with relative distance scheme. In terms of adjusted rand index, rand index, mirkins index and huberts index, better performance is achieved by proposed system when compared to the existing system which is shown by experimental result.

Is chiropterophily an evolutionary dead end? A phylogenetic test in the pantropical genus Parkia (Leguminosae, Caesalpinioideae, mimosoid clade)

Background and aims – Pollination systems often reflect adaptations to specific groups of pollinators, and these morphological specialisations have been important in the diversification of the angiosperms. Here, we study the evolution of the capitulum and pollination system in the pantropical genus Parkia, which comprises 35 species of trees distributed largely in the forests of South and Central America, Africa, Madagascar, and the Indo-Pacific. The flowers are grouped into capitula that are composed of one, two, or three distinct morphological types, and are principally pollinated either by insects or by bats. Material and methods – Using BEAST, we estimated the ages of nodes in a phylogeny based on four chloroplast regions (matK, trnL, psbA-trnH, and rps16-trnQ) and the nuclear region ITS/18S/26S. This analysis also enabled us to reconstruct the ancestral state of the capitulum and hence infer the ancestral pollination system. Euclidean distance-based cluster analysis was performed to determine which characters are consistently related to a specific pollination system.Key results – Our results indicate that the ancestral capitulum in the genus had three types of flowers and a morphology associated with bat-pollination in both the Paleotropics and Neotropics. In one derived Neotropical clade, the number of floral types in each capitulum was reduced to two (capitulum also bat-pollinated) or one (insect-pollinated). Thus, entomophily, as seen in some Neotropical species of Parkia, has been derived from a bat-pollinated ancestor. Cluster analysis showed that the floral characters were mostly consistent with pollination systems.Conclusion – Chiropterophily is not an evolutionary dead end in Parkia because during the evolutionary history of the genus there has been at least one transition to entomophily. Parkia provides a unique example of evolutionary transitions from chiropterophily to entomophily in a pantropical genus of trees.

Optimize TOD Time-Division with Dynamic Time Warping Distance-based Non-Hierarchical Cluster Analysis

Optimize TOD Time-Division with Dynamic Time Warping Distance-based Non-Hierarchical Cluster Analysis

Precision Autism: Genomic Stratification of Disorders Making Up the Broad Spectrum May Demystify Its "Epidemic Rates".

In the last decade, Autism has broadened and often shifted its diagnostics criteria, allowing several neuropsychiatric and neurological disorders of known etiology. This has resulted in a highly heterogeneous spectrum with apparent exponential rates in prevalence. I ask if it is possible to leverage existing genetic information about those disorders making up Autism today and use it to stratify this spectrum. To that end, I combine genes linked to Autism in the SFARI database and genomic information from the DisGeNET portal on 25 diseases, inclusive of non-neurological ones. I use the GTEx data on genes’ expression on 54 human tissues and ask if there are overlapping genes across those associated to these diseases and those from SFARI-Autism. I find a compact set of genes across all brain-disorders which express highly in tissues fundamental for somatic-sensory-motor function, self-regulation, memory, and cognition. Then, I offer a new stratification that provides a distance-based orderly clustering into possible Autism subtypes, amenable to design personalized targeted therapies within the framework of Precision Medicine. I conclude that viewing Autism through this physiological (Precision) lens, rather than viewing it exclusively from a psychological behavioral construct, may make it a more manageable condition and dispel the Autism epidemic myth.

Iterative learning-based many-objective history matching using deep neural network with stacked autoencoder

This paper presents an innovative data-integration that uses an iterative-learning method, a deep neural network (DNN) coupled with a stacked autoencoder (SAE) to solve issues encountered with many-objective history matching. The proposed method consists of a DNN-based inverse model with SAE-encoded static data and iterative updates of supervised-learning data are based on distance-based clustering schemes. DNN functions as an inverse model and results in encoded flattened data, while SAE, as a pre-trained neural network, successfully reduces dimensionality and reliably reconstructs geo-models. The iterative-learning method can improve the training data for DNN by showing the error reduction achieved with each iteration step. The proposed workflow shows the small mean absolute percentage error below 4% for all objective functions, while a typical multi-objective evolutionary algorithm fails to significantly reduce the initial population uncertainty. Iterative learning-based many-objective history matching estimates the trends in water cuts that are not reliably included in dynamic-data matching. This confirms the proposed workflow constructs more plausible geo-models. The workflow would be a reliable alternative to overcome the less-convergent Pareto-based multi-objective evolutionary algorithm in the presence of geological uncertainty and varying objective functions.

A clustering procedure for mixed-type data to explore ego network typologies: an application to elderly people living alone in Italy

The analysis of ego networks has attracted a great attention recently and found application in many areas of the social sciences. In particular, the identification of network typologies has become a crucial task and a powerful tool to capture aspects of the social space or personal community in which people are embedded. In this work, we propose a distance-based clustering procedure to identify homogeneous groups of ego networks that are only described by a small number of compositional variables. The proposed approach is motivated by the empirical study of ego networks of contacts extracted from the “Family and Social Subjects” (FSS) Survey conducted by the Italian National Statistical Institute in 2016, which is not specifically oriented to network analysis. We focus on elderly respondents living alone, which can be regarded as a vulnerable category, with the aim to describe their network of contacts. First, mining relational information in FSS data, we derive the ego networks of respondents. Then, we develop a methodology for coping with the presence of heterogeneous data and small amount of information from a network perspective. To this aim, we introduce a dissimilarity measure for mixed-type data, and exploit hierarchical clustering for grouping ego networks according to their composition. In doing so, we intend to make our approach applicable to various surveys.

Distance-based clustering challenges for unbiased benchmarking studies

Benchmark datasets with predefined cluster structures and high-dimensional biomedical datasets outline the challenges of cluster analysis: clustering algorithms are limited in their clustering ability in the presence of clusters defining distance-based structures resulting in a biased clustering solution. Data sets might not have cluster structures. Clustering yields arbitrary labels and often depends on the trial, leading to varying results. Moreover, recent research indicated that all partition comparison measures can yield the same results for different clustering solutions. Consequently, algorithm selection and parameter optimization by unsupervised quality measures (QM) are always biased and misleading. Only if the predefined structures happen to meet the particular clustering criterion and QM, can the clusters be recovered. Results are presented based on 41 open-source algorithms which are particularly useful in biomedical scenarios. Furthermore, comparative analysis with mirrored density plots provides a significantly more detailed benchmark than that with the typically used box plots or violin plots.

Ensembles of localised models for time series forecasting

With large quantities of data typically available nowadays, forecasting models that are trained across sets of time series, known as Global Forecasting Models (GFM), are regularly outperforming traditional univariate forecasting models that work on isolated series. As GFMs usually share the same set of parameters across all time series, they often have the problem of not being localised enough to a particular series, especially in situations where datasets are heterogeneous. We study how ensembling techniques can be used with generic GFMs and univariate models to solve this issue. Our work systematises and compares relevant current approaches, namely clustering series and training separate submodels per cluster, the so-called ensemble of specialists approach, and building heterogeneous ensembles of global and local models. We fill some gaps in the existing GFM localisation approaches, in particular by incorporating varied clustering techniques such as feature-based clustering, distance-based clustering and random clustering, and generalise them to use different underlying GFM model types. We then propose a new methodology of clustered ensembles where we train multiple GFMs on different clusters of series, obtained by changing the number of clusters and cluster seeds. Using Feed-forward Neural Networks, Recurrent Neural Networks, and Pooled Regression models as the underlying GFMs, in our evaluation on eight publicly available datasets, the proposed models are able to achieve significantly higher accuracy than baseline GFM models and univariate forecasting methods.

Evaluation of Assam rice (Oryza sativa) genotypes for nitrogen use efficiency

We assessed genetic diversity in 26 rice genotypes including 21 landraces and five improved varieties grown in three different nitrogen doses, viz. 0, 40, and 60 kg/ha during kharif 2014-15, and identified superior nitrogen use efficient genotypes for lowland rice ecosystem. The presence of significant genotypic differences was registered for 12 quantitative traits of nitrogen use efficiency, grain yield, and its related traits. Mahalanobis distance-based clustering grouped the genotypes into seven clusters. The maximum contribution towards divergence was due to flag leaf area followed by physiological nitrogen use efficiency and grain yield/plant. Principal component analysis performed on the 12 quantitative traits revealed that the first four principal components accounted for 83.57% variance among the genotypes. Grain yield/plant (0.43), nitrogen utilization efficiency (0.39), and nitrogen harvest index (0.37) accounted most for principal component 1. The genotypes namely Lothabor, Betguti Sali, and Luhasali were found to be suitable for the lowland rice ecosystem.

Genome-Wide Assessment of Population Structure and Genetic Diversity of the Global Finger Millet Germplasm Panel Conserved at the ICRISAT Genebank.

Finger millet [Eleusine coracana (L.) Gaertn.] is an important climate-resilient nutrient-dense crop grown as a staple food grain in Asia and Africa. Utilizing the full potential of the crop mainly depends on an in-depth exploration of the vast diversity in its germplasm. In this study, the global finger millet germplasm diversity panel of 314 accessions was genotyped, using the DArTseq approach to assess genetic diversity and population structure. We obtained 33,884 high-quality single nucleotide polymorphism (SNP) markers on 306 accessions after filtering. Finger millet germplasm showed considerable genetic diversity, and the mean polymorphic information content, gene diversity, and Shannon Index were 0.110, 0.114, and 0.194, respectively. The average genetic distance of the entire set was 0.301 (range 0.040 – 0.450). The accessions of the race elongata (0.326) showed the highest average genetic distance, and the least was in the race plana (0.275); and higher genetic divergence was observed between elongata and vulgaris (0.320), while the least was between compacta and plana (0.281). An average, landrace accessions had higher gene diversity (0.144) and genetic distance (0.299) than the breeding lines (0.117 and 0.267, respectively). A similar average gene diversity was observed in the accessions of Asia (0.132) and Africa (0.129), but Asia had slightly higher genetic distance (0.286) than African accessions (0.276), and the distance between these two regions was 0.327. This was also confirmed by a model-based STRUCTURE analysis, genetic distance-based clustering, and principal coordinate analysis, which revealed two major populations representing Asia and Africa. Analysis of molecular variance suggests that the significant population differentiation was mainly due to within individuals between regions or between populations while races had a negligible impact on population structure. Finger millet diversity is structured based on a geographical region of origin, while the racial structure made negligible contribution to population structure. The information generated from this study can provide greater insights into the population structure and genetic diversity within and among regions and races, and an understanding of genomic-assisted finger millet improvement.

The rhythms of rhythm

The low frequency (LF) spectral analysis or ‘rhythm spectrum’ approach to the quantitative analysis and comparison of speech rhythms is extended beyond syllable or word rhythms to ‘rhetorical rhythms’ in read-aloud narratives, in a selection of exploratory scenarios, with the aim of developing a unified theory of speech rhythms. Current methodologies in the field are first discussed, then the choice of data is motivated and themodulation-theoreticrhythm spectrum and rhythm spectrogram approach is applied to the amplitude modulation (AM) and frequency modulation (FM) of speech. New concepts ofrhythm formant,rhythm spectrogramandrhythm formant trajectoryare introduced in theRhythm Formant Theory(RFT) framework with its associated methodologyRhythm Formant Analysis(RFA) in order to capture second order regularities in the temporal variation of rhythms. The interaction of AM and FM rhythm factors is explored, contrasting English with Mandarin Chinese. The LF rhythm spectrogram is introduced in order to recover temporal information about long-term rhythms, and to investigate the configurative function of rhythm. The trajectory of highest magnitude frequencies through the component spectra of the LF spectrogram is extracted and applied in classifying readings in different languages and individual speaking styles using distance-based hierarchical clustering, and the existence of long-term second order ‘rhythms of rhythm’ in long narratives is shown. In the conclusion, pointers are given to the extension of this exploratory RFT rhythm approach for future quantitative confirmatory investigations.

An Effective Analysis of Data Clustering using Distance-based K- Means Algorithm

Real-world data sets are regularly provides different and complementary features of information in an unsupervised way. Different types of algorithms have been proposed recently in the genre of cluster analysis. It is arduous to the user to determine well in advance which algorithm would be the most suitable for a given dataset. Techniques with respect to graphs are provides excellent results for this task. However, the existing techniques are easily vulnerable to outliers and noises with limited idea about edges comprised in the tree to divide a dataset. Thus, in some fields, the necessacity for better clustering algorithms it uses robust and dynamic methods to improve and simplify the entire process of data clustering has become an important research field. In this paper, a novel distance-based clustering algorithm called the entropic distance based K-means clustering algorithm (EDBK) is proposed to eradicate the outliers in effective way. This algorithm depends on the entropic distance between attributes of data points and some basic mathematical statistics operations. In this work, experiments are carry out using UCI datasets showed that EDBK method which outperforms the existing methods such as Artificial Bee Colony (ABC), k-means.

Characterization of Oilseed Crop Noug (Guizotia abyssinica) Using Agro-Morphological Traits

Noug (Guizotia abyssinica) is an outcrossing oilseed crop that serves as a source of edible oil and other nutrients although its seed yield is generally low. The analysis of agro-morphological traits in relation to seed and oil yields is important for improving the productivity of this crop. The present study aimed at assessing the variation and heritability of quantitative (10) and qualitative (6) traits in noug based on 60 landrace accessions collected from wide geographic area in Ethiopia. The field trial was conducted at two sites in Ethiopia using a square lattice design. The analysis of variance revealed significant variation (p < 0.05) among these accessions. The highest broad-sense heritability (H2) was recorded for days to 10% flowering (DTF10; 85.4%), whereas number of capitula per plant (NCPP) and number of seeds per capitulum (NSPC) showed medium heritability (H2 = 38.5% and 31.6%, respectively). NCPP and NSPC showed a highly significant (p < 0.01) positive and negative genotypic correlation with days to flowering, respectively. These three traits showed very low genotypic coefficient of variation (<1%). In the case of qualitative traits, small capitulum, large flower, green leaf, green stem and course leaf margin showed significant association with higher number of seeds per plant (NSPP). Euclidean distance-based cluster analysis revealed that the clustering pattern of the accessions poorly correlates with the geographic distance between sample collecting sites. Similarly, no clear clustering pattern of accessions was revealed by principal component analysis (PCA) that explained 66.3% and 53.6% of the total variation of quantitative traits and qualitative traits, respectively. The oil content of these accessions was previously investigated and accessions with high oil content show large differences in terms of days to flowering, NSPP and thousand seed weight (TSW). Among the accessions included in this research, Hr_B21; Gj_C17, Sh_I4 and Gr_F15 Gj_G18 and Tg-R13 are top ranking, as they have at least one the following highly desirable traits: early maturity, high oil content, NSPP and TSW. Hence, crossbreeding of their selected genotypes would lead to the development of new cultivars that combine early maturity and both high seed and oil yields.

A co-fractionation mass spectrometry-based prediction of protein complex assemblies in the developing rice aleurone-subaleurone.

Multiprotein complexes execute and coordinate diverse cellular processes such as organelle biogenesis, vesicle trafficking, cell signaling, and metabolism. Knowledge about their composition and localization provides useful clues about the mechanisms of cellular homeostasis and system-level control. This is of great biological importance and practical significance in heterotrophic rice (Oryza sativa) endosperm and aleurone-subaleurone tissues, which are a primary source of seed vitamins and stored energy. Dozens of protein complexes have been implicated in the synthesis, transport, and storage of seed proteins, lipids, vitamins, and minerals. Mutations in protein complexes that control RNA transport result in aberrant endosperm with shrunken and floury phenotypes, significantly reducing seed yield and quality. The purpose of this study was to broadly predict protein complex composition in the aleurone-subaleurone layers of developing rice seeds using co-fractionation mass spectrometry. Following orthogonal chromatographic separations of biological replicates, thousands of protein elution profiles were subjected to distance-based clustering to enable large-scale multimerization state measurements and protein complex predictions. The predicted complexes had predicted functions across diverse functional categories, including novel heteromeric RNA binding protein complexes that may influence seed quality. This effective and open-ended proteomics pipeline provides useful clues about system-level posttranslational control during the early stages of rice seed development.

Distance-based K-Means Clustering Algorithm for Anomaly Detection in Categorical Datasets

Distance-based K-Means Clustering Algorithm for Anomaly Detection in Categorical Datasets