Model-based Clustering Method Research Articles

#782 Latent profiles of patients undergoing maintenance hemodialysis based on hemodynamic indicators

Abstract Background and Aims Both intradialytic hypotension and intradialytic hypertension are conditions occurring frequently during hemodialysis and are associated with unfavorable clinical outcomes. However, they remain incompletely understood and poorly defined in the literature. Since intradialytic parameters are now captured automatically in electronic medical records in many institutions, machine learning could emerge as an innovative tool for hemodialysis clinical research. Latent Profile Analysis is a model-based clustering method. It allows to identify hidden subpopulations derived from a set of observed indicator variables, maximizing between-group differences and within-group similarities. Method The objectives of this study were to identify latent profiles using a set of hand-crafted indicators and to interpret these subgroups of patients in a clinically meaningful manner. We conducted a retrospective single-center cohort study of adult patients receiving hemodialysis between January 2017 and December 2022 in Montreal. Given the highly heterogeneous nature of this population, we only enrolled incidental patients. The data were extracted from the NephroCare database, which is the clinical information system used by dialysis teams in our center. Sixteen indicators were derived from time series of blood pressure and heart rate measurements, capturing various patient-centric aspects of intradialytic hemodynamics, including trends, zeniths, nadirs, and time-related features. Latent Profile Analysis involved fitting a series of models, and the best model was selected based on fit indices, evaluation metrics, and clinical interpretation of the distribution of indicator within identified profiles. The primary analysis used complete-case data, and a sensitivity analysis assessed the impact of missing data through multiple imputation. Internal model validation was conducted using k-fold cross-validation. Results The selected model consisted of four profiles with varying variance and null covariance across indicators. Fig. 1 presents radar plots summarizing the identified profiles. Profile 1 included patients with prolonged early nadirs and a nadir-to-zenith transition pattern at the session level. Profile 2 was characterized by frequent nadirs and zeniths without a specific timing preference. Profile 3 comprised patients with infrequent blood pressure variations altogether. Profile 4 included patients with frequent early nadirs but no zeniths. The model demonstrated good generalization to unseen cases and exhibited relative robustness to missing data. Conclusion In this preliminary study, we demonstrated that Latent Profile Analysis can identify clinically meaningful subpopulations of hemodialysis patients based on hemodynamic indicators. The association between the identified profiles and hard clinical endpoints remains to be established.

FPDclustering: a comprehensive R package for probabilistic distance clustering based methods

AbstractData clustering has a long history and refers to a vast range of models and methods that exploit the ever-more-performing numerical optimization algorithms and are designed to find homogeneous groups of observations in data. In this framework, the probability distance clustering (PDC) family methods offer a numerically effective alternative to model-based clustering methods and a more flexible opportunity in the framework of geometric data clustering. Given nJ-dimensional data vectors arranged in a data matrix and the number K of clusters, PDC maximizes the joint density function that is defined as the sum of the products between the distance and the probability, both of which are measured for each data vector from each center. This article shows the capabilities of the PDC family, illustrating the package .

Parsimonious Bayesian model-based clustering with dissimilarities

Clustering techniques are used to group observations and discover interesting patterns within data. Model-based clustering is one such method that is often an attractive choice due to the specification of a generative model for the given data and the ability to calculate model-selection criteria, which is in turn used to select the number of clusters. However, when only distances between observations are available, model-based clustering can no longer be used, and heuristic algorithms without the aforementioned advantages are usually used instead. As a solution, Oh and Raftery (2007) suggest a Bayesian model-based clustering method (named BMCD) that only requires a dissimilarity matrix as input, while also accounting for the measurement error that may be present within the observed data. In this paper, we extend the BMCD framework by proposing several additional models, alternative model selection criteria, and strategies for reducing computing time of the algorithm. These extensions ensure that the algorithm is effective even in high-dimensional spaces and provides a wide range of choices to the practitioner that can be used with a variety of data. Additionally, a publicly available software implementation of the algorithm is provided as a package in the R programming language.

A model-based clustering via mixture of hierarchical models with covariate adjustment for detecting differentially expressed genes from paired design

The causes of many complex human diseases are still largely unknown. Genetics plays an important role in uncovering the molecular mechanisms of complex human diseases. A key step to characterize the genetics of a complex human disease is to unbiasedly identify disease-associated gene transcripts on a whole-genome scale. Confounding factors could cause false positives. Paired design, such as measuring gene expression before and after treatment for the same subject, can reduce the effect of known confounding factors. However, not all known confounding factors can be controlled in a paired/match design. Model-based clustering, such as mixtures of hierarchical models, has been proposed to detect gene transcripts differentially expressed between paired samples. To the best of our knowledge, no model-based gene clustering methods have the capacity to adjust for the effects of covariates yet. In this article, we proposed a novel mixture of hierarchical models with covariate adjustment in identifying differentially expressed transcripts using high-throughput whole-genome data from paired design. Both simulation study and real data analysis show the good performance of the proposed method.

Cinderella’s Family Tree. A Phylomemetic Case Study of ATU 510/511

Abstract This case study contributes to recent attempts to apply “phylomemetic” methods derived from computational biology to oral traditions, where the aim is to trace the mutation and diversification of folk narratives as they get passed on from generation to generation and spread from society to society. Our study focuses on one of the most famous and widespread tales in the folktale record: Cinderella. Thousands of Cinderella-like stories have been documented from around the world, which folklorists have attempted to classify into different “types” representing distinct, though related, international traditions. The most comprehensive of Cinderella typologies was developed by Anna Birgitta Rooth (1951), who divided the tales into five principal types: A, B, AB, BI and C, and suggested several hypotheses pertaining to their origins and relationships to one another. Here, we test Rooth’s theories on a sample of 266 versions of Cinderella using Bayesian phylogenetic inference, phylogenetic networks (NeighborNet) and a model-based clustering method that was originally designed to elicit population structure from multi-locus genotype data (implemented in the program STRUCTURE). Our results find varying levels of support for the types identified by Rooth, and suggest that mixing among traditions was widespread, especially in Type AB tales. Despite these complexities, it was still possible to delineate and quantify the influence of distinct ancestral sources on the variation observed in contemporary versions of Cinderella. Our study highlights the value and versatility of phylomemetic methods in uncovering the historical relationships among types and sub-types of international folktale, as well as the evolutionary processes that have shaped them.

A Proposed Clustering Algorithm for Efficient Clustering of High-Dimensional Data

To partition transaction data values, clustering algorithms are used. To analyse the relationships between transactions, similarity measures are utilized. Similarity models based on vectors perform well with low-dimensional data. High-dimensional data values are clustered using subspace clustering techniques. Clustering high-dimensional data is difficult due to the curse of dimensionality. Projective clustering seeks out projected clusters in subsets of a data space's dimensions. In high-dimensional data space, a probability model represents predicted clusters. A model-based fuzzy projection clustering method to find clusters with overlapping boundaries in different projection subspaces. The system employs the Model Based Projective Clustering (MPC) method. To cluster high-dimensional data, projective clustering algorithms are used. A subspace clustering technique is the model-based projective clustering algorithm. Similarity analysis use non-axis-subspaces. Anomaly transactions are segmented using projected clusters. The suggested system is intended to cluster objects in high-dimensional spaces. The similarity analysis includes non-access subspaces. The clustering procedure validates anomaly data values with similarity. The subspace selection procedure has been optimized. A subspace clustering approach is the model-based projective clustering algorithm. Similarity analysis use non-axis-subspaces. Anomaly transactions are segmented using projected clusters. The suggested system is intended to cluster objects in high-dimensional spaces. The similarity analysis includes non-access subspaces. The clustering procedure validates anomaly data values with similarity. The subspace selection procedure has been improved.

Time series clustering with an EM algorithm for mixtures of linear Gaussian state space models

In this paper, we consider the task of clustering a set of individual time series while modeling each cluster, that is, model-based time series clustering. The task requires a parametric model with sufficient flexibility to describe the dynamics in various time series. To address this problem, we propose a novel model-based time series clustering method with mixtures of linear Gaussian state space models, which have high flexibility. The proposed method uses a new expectation-maximization algorithm for the mixture model to estimate the model parameters, and determines the number of clusters using the Bayesian information criterion. Experiments on a simulated dataset demonstrate the effectiveness of the method in clustering, parameter estimation, and model selection. The method is applied to real datasets commonly used to evaluate time series clustering methods. Results showed that the proposed method produces clustering results that are as accurate or more accurate than those obtained using previous methods.

Robust TPMB Filtering Using Sensors With Limited Sensing Range Under Nonuniform Clutter Background

This study proposes a novel solution using sensors with limited sensing range to estimate the random appearance or disappearance of multi-object trajectories with unknown object detection profiles under nonuniform clutter background. Specifically, assuming that no objects occur within the sensors’ observable area, the unknown clutter intensity involving the average false alarms per scan and nonuniform clutter density is first estimated by adopting the joint combination of maximum likelihood (ML) and model-based clustering methods. Then, the unknown object detection profile can be calculated by marginalizing the fluctuating signal-to-noise ratio (SNR) in a random manner. Further, the TPMB filter is utilized to estimate trajectories from the first principle, after which the estimated parameters of clutter intensity and the detection probability are fed to the TPMB filtering, thereby improving the performance on completeness and robustness of estimation trajectories. Simulation and experimental results demonstrated that our proposed solution exhibits excellent tracking accuracy compared with state-of-the-art solutions involving robust MS generalized LMB (R-MS-GLMB) and dynamic parameter GLMB (DP-GLMB) filters.

Poisson hurdle model-based method for clustering microbiome features.

High-throughput sequencing technologies have greatly facilitated microbiome research and have generated a large volume of microbiome data with the potential to answer key questions regarding microbiome assembly, structure and function. Cluster analysis aims to group features that behave similarly across treatments, and such grouping helps to highlight the functional relationships among features and may provide biological insights into microbiome networks. However, clustering microbiome data are challenging due to the sparsity and high dimensionality. We propose a model-based clustering method based on Poisson hurdle models for sparse microbiome count data. We describe an expectation-maximization algorithm and a modified version using simulated annealing to conduct the cluster analysis. Moreover, we provide algorithms for initialization and choosing the number of clusters. Simulation results demonstrate that our proposed methods provide better clustering results than alternative methods under a variety of settings. We also apply the proposed method to a sorghum rhizosphere microbiome dataset that results in interesting biological findings. R package is freely available for download at https://cran.r-project.org/package=PHclust. Supplementary data are available at Bioinformatics online.

Piper section Muldera (Piperaceae) of the Malay Peninsula: variation and species limits

Abstract Species limits in Piper spp. have been unstable, with specimens often assigned to different species by different botanists. Using a strongly supported clade of Asian Piper section Muldera, I investigated species boundaries in the eight Malay Peninsula species using morphological, anatomical and molecular approaches. I assessed morphological variation using principal component analysis (PCA) and a model-based clustering method. Herbarium specimens and samples collected in the field were examined, and the latter were also used to examine variation in vegetative anatomy. Potential evidence of reciprocal monophyly of species came from the molecular phylogenetic analyses of nuclear ITS and g3pdh and plastid trnL-F. Morphological analyses did not lend support for any of the hypothesized species from the Malay Peninsula. Nuclear phylogenetic trees supported a combination of Piper schizonephros, Piper polygynum and Piper maingayi as a single species that might also have a distinct distribution of calcium oxalate crystals in the mesophyll of the leaf and a prominently raised leaf midrib raise. Nevertheless, shared characters between this species and some, but not all, of the other morphogroups, and paraphyly of other morphogroups in the phylogenetic analyses, might suggest hybridization or lateral gene transfer among species of the section Muldera. In-depth population studies coupled with examination of morphological variation for all species of the section Muldera, including those from outside the Malay Peninsula, are necessary to disentangle the species limits and ultimately the evolutionary history in this group.

House Prices Segmentation Using Gaussian Mixture Model-Based Clustering

House is a place for humans to live and a main necessity for humans. For years, the need for houses is increasing and varied so that it affects the selling price of the house. Therefore, more research is needed to learn about the selling price of houses. This research is only focusing on house price segmentation in DKI Jakarta using the Gaussian Mixture Model-Based Clustering Method with the Expectation-Maximization algorithm. The goal of this research is to make a house price segmentation model so that we can obtain useful information for the potential buyer. Clustering with GMM utilize the log-likelihood function to optimize the GMM parameters. The result of this research is houses in DKI Jakarta can be segmented into 3 different clusters. The first cluster is for the low-profile houses. The second cluster is for the mid-profile houses. The third cluster is for the high-profile houses. The silhouette score that was produced by the clustering method is 0.60866 meaning that this score is quite good because it’s close to a value of 1.

Semiparametric finite mixture of regression models with Bayesian P-splines

AbstractMixture models provide a useful tool to account for unobserved heterogeneity and are at the basis of many model-based clustering methods. To gain additional flexibility, some model parameters can be expressed as functions of concomitant covariates. In this Paper, a semiparametric finite mixture of regression models is defined, with concomitant information assumed to influence both the component weights and the conditional means. In particular, linear predictors are replaced with smooth functions of the covariate considered by resorting to cubic splines. An estimation procedure within the Bayesian paradigm is suggested, where smoothness of the covariate effects is controlled by suitable choices for the prior distributions of the spline coefficients. A data augmentation scheme based on difference random utility models is exploited to describe the mixture weights as functions of the covariate. The performance of the proposed methodology is investigated via simulation experiments and two real-world datasets, one about baseball salaries and the other concerning nitrogen oxide in engine exhaust.

Local node feature modeling for edge computing based on network embedding in dynamic networks

In a dynamic network, the characteristics of local nodes include first and higher-order proximity among the nodes as well as different attributes attached to each node. This complexity impose significant challenge for dynamic network modeling. As a result, few dynamic network studies have considered high-order proximity among local nodes. In this paper, we adopt the network embedding method to map high-order proximity of local nodes into low-dimensional, dense and real-valued vectors. Morevoer, we incorporate it into a model-based evolutionary clustering method through regularity conditions. Such a unified framework can increase the effectiveness and robustness of dynamic community detection while pertaining a good explanatory and visualization ability. Experiments based on synthetic and real world data sets show that our model can produce better community detection results than other popular models such as DECS and Genlouvain in dense networks. This result is consistent with the advantage of network embedding method in dense networks.

Hierarchical Means Clustering

In the cluster analysis literature, there are several partitioning (non-hierarchical) methods for clustering multivariate objects based on model estimation. Distinct to these methods is the use of a system of n nested statistical models and the optimization of a loss function to best-fit a clustering model to observed data. Many hierarchical clustering methods are not model-based where hierarchy is obtained using a divisive or agglomerative greedy procedure. This paper aims to fill this gap by proposing a novel hierarchical cluster analysis methodology called Hierarchical Means Clustering. HMC produces a set of nested partitions with a centroid-based model estimated via least-squares by minimizing the total within-cluster deviance of the n partitions in the hierarchy. Hierarchical Means Clustering produces a hierarchy formed by n-1 nested partitions from 2 to n clusters with minimal total cluster deviance. Six real data examples are featured, and key links to k-means, Ward’s method, Bisecting k-means and model-based hierarchical agglomerative clustering methods are discussed.

Classification of ring-spun yarns using cluster analysis

The aim of this study is to classify ring-spun yarns according to their unevenness, imperfections, and hairiness parameters using cluster analysis. The mentioned features of ring-spun yarns are measured for five different ranks. Five ranks of ring-spun yarns including compact and conventional as well as combed and carded types are chosen and produced. In the modeling section, the model-based clustering method was applied as a strong method based on the distribution of each variable. In order to select the best fit and to find out the final clustering, bayesian information criterion (BIC) is applied. According to the results of modeling, five ranks of selected ring-spun yarns are classified in four clusters and the acceptable agreement is measured according to Cohen’s kappa method. The highest value for Kappa represents a high agreement to match between the clustering result and the real rank.

DIRICHLET-TREE MULTINOMIAL MIXTURES FOR CLUSTERING MICROBIOME COMPOSITIONS.

Studying the human microbiome has gained substantial interest in recent years, and a common task in the analysis of these data is to cluster microbiome compositions into subtypes. This subdivision of samples into subgroups serves as an intermediary step in achieving personalized diagnosis and treatment. In applying existing clustering methods to modern microbiome studies including the American Gut Project (AGP) data, we found that this seemingly standard task, however, is very challenging in the microbiome composition context due to several key features of such data. Standard distance-based clustering algorithms generally do not produce reliable results as they do not take into account the heterogeneity of the cross-sample variability among the bacterial taxa, while existing model-based approaches do not allow sufficient flexibility for the identification of complex within-cluster variation from cross-cluster variation. Direct applications of such methods generally lead to overly dispersed clusters in the AGP data and such phenomenon is common for other microbiome data. To overcome these challenges, we introduce Dirichlet-tree multinomial mixtures (DTMM) as a Bayesian generative model for clustering amplicon sequencing data in microbiome studies. DTMM models the microbiome population with a mixture of Dirichlet-tree kernels that utilizes the phylogenetic tree to offer a more flexible covariance structure in characterizing within-cluster variation, and it provides a means for identifying a subset of signature taxa that distinguish the clusters. We perform extensive simulation studies to evaluate the performance of DTMM and compare it to state-of-the-art model-based and distance-based clustering methods in the microbiome context, and carry out a validation study on a publicly available longitudinal data set to confirm the biological relevance of the clusters. Finally, we report a case study on the fecal data from the AGP to identify compositional clusters among individuals with inflammatory bowel disease and diabetes. Among our most interesting findings is that enterotypes (i.e., gut microbiome clusters) are not always defined by the most dominant species as previous analyses had assumed, but can involve a number of less abundant OTUs, which cannot be identified with existing distance-based and method-based approaches.

Model-based clustering (MBC) for road data via multivariate mixture of normal distributions and factor analysis (FA)

In road traffic, there are many influential parameters that need to be analysed. For a long time, the statistical methods have been used for analysing, planning and preventing accidents in advanced countries. In this study, multivariate normal distribution was applied to a finite mixture model for clustering roads. In order to solve the problems of road traffic safety and road accidents, a new FA (Factor analysis)-MBC (Model based clustering) method was introduced based on the math theory of FA-MBC. First, the FA method was used to reduce the dimensions of the features, and then our proposed method (MBC) was applied to cluster the road data gathered for Tehran Province, Iran. The results show that our model has a good performance in dealing with data that may have high-dimensional features.

ZINB-Based Graph Embedding Autoencoder for Single-Cell RNA-Seq Interpretations

Single-cell RNA sequencing (scRNA-seq) provides high-throughput information about the genome-wide gene expression levels at the single-cell resolution, bringing a precise understanding on the transcriptome of individual cells. Unfortunately, the rapidly growing scRNA-seq data and the prevalence of dropout events pose substantial challenges for cell type annotation. Here, we propose a single-cell model-based deep graph embedding clustering (scTAG) method, which simultaneously learns cell–cell topology representations and identifies cell clusters based on deep graph convolutional network. scTAG integrates the zero-inflated negative binomial (ZINB) model into a topology adaptive graph convolutional autoencoder to learn the low-dimensional latent representation and adopts Kullback–Leibler (KL) divergence for the clustering tasks. By simultaneously optimizing the clustering loss, ZINB loss, and the cell graph reconstruction loss, scTAG jointly optimizes cluster label assignment and feature learning with the topological structures preserved in an end-to-end manner. Extensive experiments on 16 single-cell RNA-seq datasets from diverse yet representative single-cell sequencing platforms demonstrate the superiority of scTAG over various state-of-the-art clustering methods.

Clustering compositional data using Dirichlet mixture model.

A model-based clustering method for compositional data is explored in this article. Most methods for compositional data analysis require some kind of transformation. The proposed method builds a mixture model using Dirichlet distribution which works with the unit sum constraint. The mixture model uses a hard EM algorithm with some modification to overcome the problem of fast convergence with empty clusters. This work includes a rigorous simulation study to evaluate the performance of the proposed method over varied dimensions, number of clusters, and overlap. The performance of the model is also compared with other popular clustering algorithms often used for compositional data analysis (e.g. KMeans, Gaussian mixture model (GMM) Gaussian Mixture Model with Hard EM (Hard GMM), partition around medoids (PAM), Clustering Large Applications based on Randomized Search (CLARANS), Density-Based Spatial Clustering of Applications with Noise (DBSCAN) etc.) for simulated data as well as two real data problems coming from the business and marketing domain and physical science domain, respectively. The study has shown promising results exploiting different distributional patterns of compositional data.

Model-based clustering of high-dimensional longitudinal data via regularization.

We propose a model-based clustering method for high-dimensional longitudinal data via regularization in this paper. This study was motivated by the Trial of Activity in Adolescent Girls (TAAG), which aimed to examine multilevel factors related to the change of physical activity by following up a cohort of 783 girls over 10 years from adolescence to early adulthood. Our goal is to identify the intrinsic grouping of subjects with similar patterns of physical activity trajectories and the most relevant predictors within each group. The previous analyses conducted clustering and variable selection in two steps, while our new method can perform the tasks simultaneously. Within each cluster, a linear mixed-effects model (LMM) is fitted with a doubly penalized likelihood to induce sparsity for parameter estimation and effect selection. The large-sample joint properties are established, allowing the dimensions of both fixed and random effects to increase at an exponential rate of the sample size, with a general class of penalty functions. Assuming subjects are drawn from a Gaussian mixture distribution, model effects and cluster labels are estimated via a coordinate descent algorithm nested inside the Expectation-Maximization (EM) algorithm. Bayesian Information Criterion (BIC) is used to determine the optimal number of clusters and the values of tuning parameters. Our numerical studies show that the new method has satisfactory performance and is able to accommodate complex data with multilevel and/or longitudinaleffects.