Model-based Cluster Analysis Research Articles

Exploring digital interactivity: a multimodal and social semiotic approach to intersemiotic construction in COVID China website

Abstract This paper explores intersemiotic construction and digital interactivity through a multimodal social semiotic lens examining the semiotic instantiations of interactive signs on the homepage of “Fighting COVID-19 the Chinese Way” (covid-19.chinadaily.com.cn), a website created to disseminate information on news and development of COVID prevention and control practices in China to contribute to global efforts to fight the pandemic. The study focuses on how digital interactivity is afforded by the COVID China website, where the interactive signs are ideationally and compositionally constructed for meaning representation and interpersonally for text-viewer relation construal. From the multimodal social semiotic perspective, systemic functional model-based cluster and intersemiotic analyses are applied to explore the visual, spatial, and linguistic features that contribute to the design and construction of interactive semiotic signs on the COVID China website and afford digital interactivity for viewers’ action potentials. This study extends the analytical focus to the semiotic instantiations of interactive signs and their intersemiotic construction process that stimulate the enabling of interactivity, instead of the interactivity per se, and demonstrates how different semiotic instantiations of interactive signs are featured and interact to afford digital interactivity. It argues for an integrated lens in analysis to look at the interactive signs not only as signs of action with action-enabling forms but also as signs of meanings that afford user-page interactivity.

Socioeconomic status and parent perceptions about the costs and benefits of youth sport.

ObjectivesDescribe what costs and benefits parents across the socioeconomic spectrum weight most heavily when making decisions about sport participation for their children.MethodCross-sectional survey of a nationally representative online panel of parents of children between the ages of 5 and 18 (n = 1025, 52% response rate). Parents rated the importance of a series of potential costs and benefits of youth sport and these responses were compared across tertiles of per capita family income. We first examined the association between family income tertiles and cost and benefit variables. Model-based cluster analysis was then used to identity homogeneous groups of responses to costs and benefits.ResultsIn all income tertiles, the top two benefits of sport were the same: having fun and being physically active. Sport as a means of keeping children out of trouble was very important for 64% of low-income parents as compared to 40% of high-income parents. Obtaining a college athletic scholarship was very important for 26% of low-income parents, as compared to 8% of high-income parents. Relative rankings of potential costs were similar by income tertile, with risk of concussion and other injury and the impact of sport on schoolwork prioritized across tertiles.ConclusionsParents prioritized fun and fitness in sport, and were concerned about injury and the impact of sport on academics. Lower income parents were the most likely to view keeping their child out of trouble, and the potential for a college athletics scholarship, as benefits of sport. Efforts to support parental decision making should be grounded in an understanding that family preferences are contextually constrained. While all parents should be appropriately informed about the potential costs and benefits they are weighting in their sports-related decision making, such family-focused efforts should be balanced with the recognition that structural change is needed to address income-related concerns about sport participation.

Whole-Genome SNP Characterisation Provides Insight for Sustainable Use of Local South African Livestock Populations.

Local cattle and sheep populations are important for animal production and food security in South Africa. These genetic resources are well adapted to the diverse climatic conditions and hold potential to be utilized in production systems subjected to climate change. The local beef breeds are well integrated into commercial livestock production systems with access to performance recording and genetic evaluations, while local sheep breeds are mainly utilised in smallholder and communal systems. The GeneSeek® Genomic Profiler™ Bovine 150 K SNP genotyping array was used to evaluate the diversity and inbreeding status of four indigenous (Boran, Drakensberger, Nguni, Tuli), two composite (Bonsmara and Beefmaster) and two exotic (SA Hereford and Charolais) beef breeds. The Illumina® Ovine 50 K SNP BeadChip was used to investigate five indigenous (Black Head Persian, Damara, Fat tail, Namaqua Afrikaner, Pedi) and three commercial (Dorper, Dohne Merino and SA Merino) populations. Although ascertainment bias was indicated by the low MAF (the autosome-wide proportion of SNPs with MAF< 0.05 ranged from 6.18 to 9.97% for cattle, and 7.59–13.81% for sheep), moderate genomic diversity was observed (mean Ho ranged from 0.296 to 0.403 for cattle, and 0.327 to 0.367 for sheep). Slightly higher levels of ROH-based inbreeding were calculated for cattle (FROH range: 0.018–0.104), than for sheep populations (FROH range: 0.002–0.031). The abundance of short ROH fragments (mean proportion of <4 Mb fragments: 0.405 for cattle, and 0.794 for sheep) indicated ancient inbreeding in both species. The eight cattle populations were categorized into indicine, taurine or Sanga subspecies based on principal component, model-based clustering and phylogenetic analyses, with high levels of admixture observed within the Drakensberger, Nguni and Tuli breeds. Within the sheep populations, a clear distinction could be seen between the dual-purpose breeds, the meat breed and the indigenous breeds. Despite directional selection practiced in the cattle breeds, genomic diversity was moderate with low inbreeding. The non-commercialized, indigenous sheep populations are more vulnerable with small effective populations. These results emphasise the value of genomic information for effective management to exploit the potential contribution of local genetic cattle and sheep resources in a changing environment.

Consensus clustering for case series identification and adverse event profiles in pharmacovigilance

ObjectiveTo describe and evaluate vigiGroup - a consensus clustering algorithm which can identify groups of individual case reports referring to similar suspected adverse drug reactions and describe associated adverse event profiles, accounting for co-reported adverse event terms. Materials and methodsConsensus clustering is achieved by grouping pairs of reports that are repeatedly placed together in the same clusters across a set of mixture model-based cluster analyses. The latter use empirical Bayes statistical shrinkage for improved performance. As baseline comparison, we considered a regular mixture model-based cluster analysis. Three randomly selected drugs in VigiBase, the World Health Organization's global database of Individual Case Safety Reports were analyzed: sumatriptan, ambroxol and tacrolimus. Clustering stability was assessed using the adjusted Rand index, ranging between −1 and +1, and clinical coherence was assessed through an intruder detection analysis. ResultsFor the three drugs considered, vigiGroup achieved stable and coherent results with adjusted Rand indices between +0.80 and +0.92, and intruder detection rates between 86% and 94%. Consensus clustering improved both stability and clinical coherence compared to mixture model-based clustering alone. Statistical shrinkage improved the stability of clusters compared to the baseline mixture model, as well as the cross-validated log-likelihood. ConclusionsThe proposed algorithm can achieve adequate stability and clinical coherence in clustering individual case reports, thereby enabling better identification of case series and associated adverse event profiles in pharmacovigilance. The use of empirical Bayes shrinkage and consensus clustering each led to meaningful improvements in performance.

How to Use Model-Based Cluster Analysis Efficiently in Person-Oriented Research.

Model-based cluster analysis (MBCA) was created to automatize the often subjective model-selection procedure of traditional explorative clustering methods. It is a type of finite mixture modelling, assuming that the data come from a mixture of different subpopulations following given distributions, typically multivariate normal. In that case cluster analysis is the exploration of the underlying mixture structure. In MBCA finding the possible number of clusters and the best clustering model is a statistical model-selection problem, where the models with differing number and type of component distributions are compared. For fitting a certain model MBCA uses a likelihood based Bayesian Information Criterion (BIC) to evaluate its appropriateness and the model with the highest BIC value is accepted as the final solution. The aim of the present study is to investigate the adequacy of automatic model selection in MBCA using BIC, and suggested alternative methods, like the Integrated Completed Likelihood Criterion (ICL), or Baudry’s method. An additional aim is to refine these procedures by using so called quality coefficients (QCs), borrowed from methodological advances within the field of exploratory cluster analysis, to help in the choice of an appropriate cluster structure (CLS), and also to compare the efficiency of MBCA in identifying a theoretical CLS with those of various other clustering methods. The analyses are restricted to studying the performance of various procedures of the type described above for two classification situations, typical in person-oriented studies: (1) an example data set characterized by a perfect theoretical CLS with seven types (seven completely homogeneous clusters) was used to generate three data sets with varying degrees of measurement error added to the original values, and (2) three additional data sets based on another perfect theoretical CLS with four types. It was found that the automatic decision rarely led to an optimal solution. However, dropping solutions with irregular BIC curves, and using different QCs as an aid in choosing between different solutions generated by MBCA and by fusing close clusters, optimal solutions were achieved for the two classification situations studied. With this refined procedure the revealed cluster solutions of MBCA often proved to be at least as good as those of different hierarchical and k-center clustering methods. MBCA was definitely superior in identifying four-type CLS models. In identifying seven-type CLS models MBCA performed at a similar level as the best of other clustering methods (such as k-means) only when the reliability level of the input variables was high or moderate, otherwise it was slightly less efficient.

Chloroplastic and nuclear diversity of endemic Prunus armeniaca L. species in the oasis agroecosystems.

Tunisia is characterized by the presence of specific seed-propagated apricot (Prunus armeniaca L.) material which is found in the oasis agroecosystems. In order to highlight the genetic diversity, population structure, and demographic history of this germplasm, 33 apricot accessions collected from six different oasis regions in southwestern Tunisia were genotyped using 24 microsatellite markers. A total number of 111 alleles was detected with an average of 4.62 alleles per locus. Bayesian model-based clustering analysis indicated four subdivisions within the collection sampled that corresponded mainly to the geographic origin of the material. The analysis of the 33 accessions using chloroplast markers allowed the identification of 32 haplotypes. Overall, the present study highlighted the high Tunisian apricot's diversity in the traditional oasis agroecosystems with low genetic differentiation. Understanding the structure of seed-propagated apricot collection is crucial for managing collections in regard to adaptive traits for Arid and Saharan climates as well as for identifying interesting genotypes that can be integrated into international coordinated actions of breeding programs.

Primary and Secondary Variants of Psychopathy in Incarcerated Youth: An Investigation with a Focus on Social Anxiety

ABSTRACT Research generally supports the notion that primary and secondary psychopathy variants are relatively distinct, with primary psychopathy reflecting a congenital affective deficit, and secondary psychopathy reflecting acquired affective disorders. However, little is known about how social anxiety might play in terms of the primary vs. secondary psychopathy typology. The main aim of the present study is to investigate whether the two distinct subtypes of psychopathy emerge using a social anxiety approach. The sample was composed of youth (N = 225, M = 16.72 years, SD = 1.44) from Portuguese juvenile detention centers. Results using model-based cluster analyses suggested the presence of two distinct groups of youth characterized by psychopathic traits that differed on social anxiety. Primary psychopathic youth presented a significantly more severe antisocial/criminal behavior profile, while secondary psychopathic youth reported a less severe antisocial/criminal behavior profile and higher levels of empathy when compared to a control cluster. Findings based on a social anxiety framework suggest that the two subtypes of psychopathy are relevant in terms of assessment and that the development of future preventive/treatment interventions should take into account this typology.

Molecular genetic survey and forensic characterization of Chinese Mongolians via the 47 autosomal insertion/deletion marker

The Mongolians are mainly distributed in the modern state of Mongolia, China, Russia, and other countries. While the historic and archaeological records of the rise and fall of the Mongol Empire are well documented, little has been known about the genetic legacy of modern Mongolian populations. Here, 611 Mongolian individuals from Hohhot, Hulunbuir, and Ordos of China were genotyped via the 47 Insertion/Deletion markers. Forensically statistical parameters indicated that this InDel system could be applied to forensic investigation in Mongolian populations. The comprehensive population comparisons indicated that targeted Mongolian populations are a homogeneous population, which kept close genetic proximity with geographically northern East Asians. The findings of the model-based clustering analysis revealed a southern East Asian-specific ancestral component, which was maximized in Hainan Li, and Mongolian populations harbored relatively less Hainan Li-related ancestry and more northern East Asian-related ancestry compared with reference Tai-Kadai, Austroasiatic and Sinitic people.

Genetic diversity and population structure of Ethiopian faba bean (Vicia faba L.) germplasm revealed by ISSR markers

Molecular characterization of faba bean accessions is important for future collection, conservation, and crop improvement programs. However, molecular genetic diversity of the Ethiopian faba bean accessions has not been comprehensively characterized. The present study was conducted to determine the genetic diversity and population structure of 90 faba bean landraces and six released varieties using nine inter-simple sequence repeat (ISSR) markers. The polymorphic information content (PIC), marker index (MI), resolving power (RP) and effective multiplex ratio (EMR) showed average values of 0.32, 5.87, 7.14, and 18.34 per primer, indicating high polymorphism values. The marker revealed average gene diversity of 0.26, ranging from 0.15 to 0.36. The degree of polymorphism among the landrace accessions ranged from 44.91% for the Sidama population to 72.46% for the East Hararghe population, with a mean of 57.27%. Analysis of molecular variance (AMOVA) indicated 99 % variation within and 1% variation among population, respectively. The present study revealed little to moderate population differentiation accompanied by high gene flow. Both distance-based and model-based cluster analysis distinguished seven distinct groups illustrating the conformity of results obtained from the study. Accessions from different administrative zones (AZs) of the country were distributed over most of the clusters; exhibiting high genetic diversity in collection sites. Therefore, further collection, conservation, and utilization programs would be recommended from AZs with high genetic diversity.

Model-Based Cluster Analysis of Democracies

We apply model-based cluster analysis to data concerning types of democracies, creating an instrument for typologies. Noting several ad- vantages of model-based clustering over traditional clustering methods, we t a normal mixture model for types of democracy in the context of the majoritarian-consensus contrast using Lijphart's (1999) data on ten vari- ables for 36 democracies. The model for the full period (1945-1996) nds four types of democracies: two types representing a majoritarian-consensus contrast, and two mixed ones lying between the extremes. The four-cluster solution shows that most of the countries have high cluster membership probabilities, and the solution is found to be quite stable with respect to possible measurement error in the variables included in the model. For the recent-period (1971-1996) data, most countries remain in the same clusters as for the full-period data.

Model-Based Cluster Analysis of Democracies

Model-Based Cluster Analysis of Democracies

A Three-Phase Fundamental Diagram from Three-Dimensional Traffic Data

This paper uses empirical traffic data collected from three locations in Europe and the US to reveal a three-phase fundamental diagram with two phases located in the uncongested regime. Model-based clustering, hypothesis testing and regression analyses are applied to the speed–flow–occupancy relationship represented in the three-dimensional space to rigorously validate the three phases and identify their gaps. The finding is consistent across the aforementioned different geographical locations. Accordingly, we propose a three-phase macroscopic traffic flow model and a characterization of solutions to the Riemann problems. This work identifies critical structures in the fundamental diagram that are typically ignored in first- and higher-order models and could significantly impact travel time estimation on highways.

Statistical analysis in support of maintaining a healthy traditional Siamese cat population

BackgroundFor many years, breeders of companion animals have applied inbreeding or line breeding to transfer desirable genetic traits from parents to their offspring. Simultaneously, this resulted in a considerable spread of hereditary diseases and phenomena associated with inbreeding depression.ResultsOur cluster analysis of kinship and inbreeding coefficients suggests that the Thai or traditional Siamese cat could be considered as a subpopulation of the Siamese cat, which shares common ancestors, although they are considered as separate breeds. In addition, model-based cluster analysis could detect regional differences between Thai subpopulations. We show that by applying optimal contribution selection and simultaneously limiting the contributions by other breeds, the genetic diversity within subpopulations can be improved.ConclusionIn principle, the European mainland Thai cat population can achieve a genetic diversity of about 26 founder genome equivalents, a value that could potentially sustain a genetically diverse population. However, reaching such a target will be difficult in the absence of a supervised breeding program. Suboptimal solutions can be obtained by minimisation of kinships within regional subpopulations. Exchanging animals between different regions on a small scale might be already quite useful to reduce the kinship, by achieving a potential diversity of 23 founder genome equivalents. However, contributions by other breeds should be minimised to preserve the original Siamese gene pool.

Model-Based Clustering, Classification, and Discriminant Analysis Using the Generalized Hyperbolic Distribution: MixGHD R package

The MixGHD package for R performs model-based clustering, classification, and discriminant analysis using the generalized hyperbolic distribution (GHD). This approach is suitable for data that can be considered a realization of a (multivariate) continuous random variable. The GHD has the advantage of being flexible due to skewness, concentration, and index parameters; as such, clustering methods that use this distribution are capable of estimating clusters characterized by different shapes. The package provides five different models all based on the GHD, an efficient routine for discriminant analysis, and a function to measure cluster agreement. This paper is split into three parts: the first is devoted to the formulation of each method, extending them for classification and discriminant analysis applications, the second focuses on the algorithms, and the third shows the use of the package on real datasets.

Self-harm among in-school and street-connected adolescents in Ghana: a cross-sectional survey in the Greater Accra region

ObjectivesTo identify the prevalence, methods, associations and reported reasons for self-harm among in-school and street-connected adolescents in Ghana.DesignA cross-sectional survey. We applied multi-level regression models and model-based cluster analysis to...

Genetic diversity of walnut (Juglans regia L.) seedlings through SSR markers in north-western Himalayan region of Jammu

screening of 25 SSR markers, revealed 23 clear and consistent amplification profiles in the entire walnut germplasm set. A total of 54 alleles were amplified by SSR primers and the number of alleles range from 2 to 3. The PIC value ranged from 0.36 to 0.68. The dendrogram classified all genotypes into two main clusters with various degrees of subclustering. Estimated genetic dissimilarity coefficient ranged from 0.36 to 0.85. Through model-based cluster analysis all genotypes were grouped into 5 genetically distinct subpopulations. The expected heterozygosity at a given locus was found to range from 0.520 to 0.5477. Similarly, population differentiation measurements (Fst) ranged from 0.2286 to 0.2909. These findings would be helpful for decision making in future walnut breeding studies, germplasm management activities to maximize genetic diversity in walnut germplasm and may also prove useful in future for conducting association mapping in walnut for different traits.

Approaches for Modeling of Intensive Longitudinal Data to Understand Cognitive Aging

Understanding age-related change in cognition and identification of pathological changes requires sensitive and valid measurement of cognitive performance across time. Technological advances, such as ambulatory assessment of cognition using smartphones, have enabled intensive longitudinal methods where data is collected with many measurements over time. Our research group has developed novel ambulatory assessments that provide reliable, sensitive, and ecologically valid measurement of cognition across multiple timescales; from momentary changes to change across years. This symposium will present a spectrum of approaches to analysis of intensive longitudinal data that can inform models of cognitive aging. All three presentations will draw on data from measurement burst studies that apply our ambulatory cognitive assessment methods in community-based samples (i.e., systematically recruited in the Bronx, New York). For each measurement burst, participants undergo assessment consisting of brief surveys and cognitive tests via smartphone, up to 7 times per day across 14 days. Oravecz et al. will discuss the application of a Bayesian multilevel implementation of the double exponential model to account for retest effects while quantifying change in peak cognitive performance across time. Kang et al., will demonstrate a growth curve modeling approach for assessing the effects of between-person variables (i.e., loneliness) on change in cognition across measurement bursts. Harrington et al., will demonstrate a model-based cluster analysis approach, leveraging ambulatory assessments of subjective and objective cognitive function to unpack latent groups as a function of age and loneliness. Measurement, Statistics, and Research Design Interest Group Sponsored Symposium.

Genetic diversity and population structure in the endangered tree Hopea hainanensis (Dipterocarpaceae) on Hainan Island, China.

Hopea hainanensis Merrill & Chun (Dipterocarpaceae) is an endangered tree species restricted to Hainan Island, China and a small part of Northern Vietnam. On Hainan Island, it is an important indicator species for tropical forests. However, because of its highly valued timber, H. hainanensis has suffered from overexploitation, leading to a sharp population decline. To facilitate the conservation of this species, genetic diversity and population structure were assessed using 12 SSR markers for 10 populations sampled across Hainan Island. Compared to non-threatened Hopea species, H. hainanensis exhibited reduced overall genetic diversity and increased population differentiation (AMOVA: FST = 0.23). Bayesian model-based clustering and principal coordinate analysis consistently assigned H. hainanensis individuals into three genetic groups, which were found to be widespread and overlapping geographically. A Mantel test found no correlation between genetic and geographical distances (r = 0.040, p = 0.418). The observed genetic structure suggests that long-distance gene flow occurred among H. hainanensis populations prior to habitat fragmentation. A recent population bottleneck was revealed, which may cause rapid loss of genetic diversity and increased differentiation across populations. Based on these findings, appropriate strategies for the long-term conservation of the endangered species H. hainanensis are proposed.

Model-based PID-5 domain clusters and levels of impairment in self and interpersonal functioning

In the DSM-5 Section III Alternative Model for Personality Disorders (AMPD), the severity of personality pathology (criterion A) and the presence of pathological personality trait domains and facets (criterion B) are assessed independently. Recent studies have challenged this assumption of independence but the interplay between criterion A and criterion B remains subject of further research. Using model-based cluster analysis on the criterion B trait domains, we compared criterion B trait domain clusters with respect to criterion A severity measures. Results revealed a six-cluster solution. Four of the six clusters represented a gradual increase in criterion B trait domains, paralleling an increase in criterion A severity. Two clusters did not follow this pattern: an Anxious-Detached type exhibiting overall high criterion A severity scores, and an emotionally stable psychopathy type exhibiting a number of low criterion A severity scores. Our findings indicate that criterion B domain clusters are informative of criterion A severity, relevant for future conceptualizations of the AMPD.

Genome-wide SNP discovery and core marker sets for DNA barcoding and variety identification in commercial tomato cultivars

Single nucleotide polymorphisms (SNPs) have been widely used as a molecular marker in crop species with advances in next-generation sequencing technology. The use of SNP markers for variety identification is a cost-effective strategy to protect breeder’s intellectual property rights. This study was conducted to identify genome-wide SNPs and develop core marker sets for assessing genetic variations in commercial tomato cultivars. A total of 10,615 confident SNPs was generated from genotyping by sequencing for 48 F1 cultivars representing four market classes (large-fruited fresh market, cherry, grape, and rootstock). Of these, 288 SNPs across 12 chromosomes were selected to genotype additional 94 F1 cultivars using the Fluidigm assay and 224 SNPs showed reliable polymorphisms in 91 F1 cultivars. Both model-based and hierarchical clustering analyses with these markers found that the large-fruited fresh market cultivars were significantly distinct from the cherry and grape cultivars. In addition, the cherry and grape cultivars showed higher levels of genetic diversity relative to the large-fruited fresh market cultivars. The 224 SNP markers were also effective in differentiating all 139 F1 cultivars and Heinz 1706 (an inbred for the tomato reference genome). Of the five subsets, the 192 and 96 markers identified all of these tomato cultivars, respectively. Furthermore, the other three subsets of 48, 24, and 12 markers showed 80.0–93.6 % of identification rates. These results demonstrate that all five subsets of SNP markers will be useful in developing a high-throughput DNA barcoding system for variety identification in commercial tomato cultivars.