Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “hierarchical clustering”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 recordsLinked to original sources
PubMed2024

Use of 3D chaos game representation to quantify DNA sequence similarity with applications for hierarchical clustering.

A 3D chaos game is shown to be a useful way for encoding DNA sequences. Since matching subsequences in DNA converge in space in 3D chaos game encoding, a DNA sequence's 3D chaos game representation can be used to compare DNA sequences without prior alignment and without truncating or padding any of the sequences. Two proposed methods inspired by shape-similarity comparison techniques show that this form of encoding can perform as well as alignment-based techniques for building phylogenetic trees. The first method uses the volume overlap of intersecting spheres and the second uses shape signatures by summarizing the coordinates, oriented angles, and oriented distances of the 3D chaos game trajectory. The methods are tested using: (1) the first exon of the beta-globin gene for 11 species, (2) mitochondrial DNA from four groups of primates, and (3) a set of synthetic DNA sequences. Simulations show that the proposed methods produce distances that reflect the number of mutation events; additionally, on average, distances resulting from deletion mutations are comparable to those produced by substitution mutations.

Animals↗
PubMed1978

Entropy in the hierarchical cluster analysis of hospitals.

A new technique integrating concepts from cluster analysis and information theory was applied to the classification of Michigan hospitals. First, a number of cost-related variables that describe the hospitals and their surroundings were used in a cluster analysis to produce a hierarchy of classifications. Then for each classification, the within-group entropy was computed for each group of hospitals and averaged over the classification. Finally, this average entropy was used as an aid to judgment in deciding which of the many classifications in the hierarchy yields the most reasonable groupings of hospitals.

Blue Cross Blue Shield Insurance Plans↗
PubMed2026

Pilot study identifying distinct circulating proteomic profiles associated with longitudinal CT-defined fibrotic and inflammatory sarcoidosis.

INTRODUCTION: Pulmonary sarcoidosis exhibits heterogeneous clinical trajectories ranging from self-limited disease resolution to chronic progressive fibrosis, yet reliable biomarkers capable of distinguishing these disease patterns remain lacking. Whether longitudinal CT-defined sarcoidosis phenotypes are associated with distinct circulating molecular signatures remains unknown. METHODS: We performed high-throughput plasma proteomics (SomaScan 11K) in participants with pulmonary sarcoidosis classified into longitudinal chest CT-defined progressive fibrosis, progressive nodular inflammatory disease, or resolving disease trajectories, along with healthy controls. CT phenotypes were assigned based on predefined longitudinal changes in reticulation, traction bronchiectasis, nodular involvement, and mediastinal lymphadenopathy across serial CT scans. One plasma sample per participant was selected from the study visit corresponding to the CT time point at which criteria for the assigned longitudinal phenotype were met. Principal component analysis, hierarchical clustering, pathway enrichment, and correlation-based analyses linking protein expression to quantitative CT features were used to evaluate whether distinct longitudinal CT phenotypes were associated with divergent proteomic signatures. RESULTS: Principal component analysis and hierarchical clustering suggested partial segregation by CT-defined phenotype. Longitudinal CT phenotypes were associated with distinct pathway-level proteomic signatures, with progressive fibrosis enriched for epithelial-mesenchymal transition signaling, and progressive nodular inflammatory disease enriched for mTORC1, MYC, oxidative phosphorylation, adipogenesis, and fatty acid metabolism pathways. Correlation analyses showed coordinated protein-expression patterns associated with fibrotic CT features and mediastinal lymph node enlargement. DISCUSSION: These findings suggest that longitudinal CT-defined fibrotic and inflammatory sarcoidosis phenotypes are associated with distinct pathway-level proteomic signatures. This pilot study provides preliminary proof-of-concept evidence that integrating longitudinal CT imaging phenotypes with plasma proteomics may serve as a framework for future mechanistic studies and biomarker discovery in pulmonary sarcoidosis.

Humans↗
PubMed2026

The potential of clustering methods for pre-test triage in sleep medicine: A systematic review.

Sleep disorders exhibit substantial heterogeneity, and traditional classifications may not fully capture clinically relevant subtypes. Clustering techniques can identify patient subgroups that improve phenotypic characterization and may support personalized management. This systematic review evaluated the application of clustering in sleep medicine, with particular focus on its potential use as a pre-test triage tool prior to formal sleep testing. PubMed/MEDLINE, Embase, Web of Science, and Scopus were searched to February 2025. Eligible studies applied clustering to classify sleep disorders in adults. Two reviewers independently conducted screening, data extraction, and risk-of-bias assessment using QUADAS-2. The protocol was registered on PROSPERO. Fifty-one studies (1983-2025) were included, predominantly focused on obstructive sleep apnea (OSA) (n = 38, 74%). Hierarchical clustering (n = 20) and K-means clustering (n = 14) were the most frequently used techniques. Internal validation was reported in only 18% of studies, and external validation was reported in only 1 study. Seven studies relied exclusively on baseline clinical, demographic, or questionnaire data, representing pre-test scenarios, whereas most incorporated polysomnography-derived variables, limiting their applicability to early clinical stratification. Hierarchical clustering was the most commonly applied method; however, the overall lack of validation limits confidence in the robustness and clinical applicability of identified phenotypes. The potential role of clustering as a pre-test triage strategy remains largely unexplored, as most studies focused on post-diagnostic phenotyping and were affected by incorporation bias. Future research should prioritize pre-test clinical variables, rigorously validate internally and externally, and adopt standardized methodological and reporting practices to facilitate clinical translation.

Humans↗
PubMed1976

Cluster analysis applied to symptom ratings of psychiatric patients: an evaluation of its predictive ability.

Rating on 39 symptoms were examined for patients admitted to the Neuropsychiatric Institute of the University of Michigan Medical Center. A detailed evaluation was made of the clusters derived by a hierarchical clustering algorithm, using complete linkage and a simple matching coefficient on the binary variables of presence or absence of symptoms. The four groups of patients suggested by the cluster analysis can be characterized as follows: (1) generalized multiplicity of symptoms; (2) capacity to cope except for orientation apart from generally held norms; (3) activity level and thought processes speeded up, intensified, and unselected; (4) inwardly punitive, slowed down and distressed. It is shown that these groups received significantly different treatment and that the effect of treatment was significantly different, while no such differences were noted for groups defined in terms of diagnoses. By means of linear discriminant functions, rules are suggested for assigning other psychiatric patients to one of these four groups.

Antipsychotic Agents↗
PubMed2026

An Integrated Machine Learning and Genomic Framework for Precise Detection of Gastric Cancer.

This study presents a novel integrative approach for the analysis of high-dimensional gene expression data, leveraging the complementary strengths of unsupervised clustering and supervised classification. Using K-means clustering, the data set is stratified into three distinct clusters, revealing intrinsic biological patterns and relationships. The resulting cluster assignments are subsequently used as pseudolabels to train machine learning models, including support vector machines, random forest, and a stacking ensemble classifier. To validate and enhance the robustness of clustering, complementary methods, such as hierarchical clustering and density-based spatial clustering of applications with noise (DBSCAN), are used, with results visualized through principal component analysis-driven dimensionality reduction. The high predictive accuracy achieved by the classifiers underlines the separability and reliability of the identified clusters. Furthermore, feature importance analysis highlighted key genetic determinants within each cluster, offering actionable insights into potential biomarkers and critical genomic features. This framework bridges the gap between exploratory unsupervised learning and predictive supervised modeling, providing a scalable and interpretable method for analyzing complex genomic data sets. Its applicability extends to biomarker discovery, patient stratification, and other precision medicine applications, emphasizing its utility in advancing genomic research and clinical practice.

Humans↗
PubMed1977

Seeing, hearing, and doing: a developmental study of memory for actions.

The ability to recall and organize actions was studied in children from 5 to 11 years in age. 8 different auditory or visual commands were successively presented for 10 trials in each modality in a free-recall task. Younger children performed fewer commands but recalled relatively more recent ones, and they showed the same degree of subjective organization and the same degree and structure of hierarchical clustering as the older children. The hierarchical structure was independent of recall, age, and modality, with the motor actions being organized by the locus of the object or instrument of the verb in the command. The difficulty of the commands was highly correlated with uncertainty of the locus of the action, that is, the number of possible arguments (objects or instruments) a verb could assume, as measured by a subsidiary experiment on 8-year-olds who were asked to name as many parts of the body upon or with which one could perform each action. Developmental differences in recall appear to rise because of primary organization (retrieval) and rehearsal strategies rather than secondary organization.

Auditory Perception↗
PubMed2026

Intraskeletal Variation in Cortical Bone Quantity in a Medieval Italian Sample: A Multivariate Exploratory Approach.

Bioarcheologists interpret skeletal health by examining variability within and between individuals. Studies of bone loss have generated contradictory and conflicting results regarding the onset and severity of age-related bone loss on a global and temporal scale, perhaps due to mismatched methodologies. Intraskeletal comparisons of bone tissue prove challenging precisely because of heterogeneous baselines in quantity and remodeling of cortical bone throughout the skeleton, as well as evolutionary histories and environmental impacts on growth and development. Here we analyze cortical bone indicators from the rib, metacarpal, and femoral cortical bone in a subset of individuals (n = 72) regions from the medieval Italian archaeological site of Pieve di Pava. To facilitate intraskeletal comparisons across elements with different biological baselines, we standardize cortical bone parameters using z-scores. Variation in relative intraskeletal cortical bone was assessed using accessible multivariate methods (principal component analysis and hierarchical cluster analysis). Results suggest an association between femoral and metacarpal cortical bone values, with stochastic trends in metacarpal and femoral relative bone quantity in relation to the rib bone quantity at the sample level. Our study demonstrates that while intraskeletal analyses are challenging, they are made more robust by synthesizing multivariate methods alongside exploratory data analysis (EDA) methods to tack between sample-level and individual-level scales and variability. Ultimately, we advocate for leveraging multivariate techniques not as a final step, but rather as a means of generating new hypotheses and challenging tendencies to a priori establish typological groups in the research process.

Skeleton↗
PubMed2026

Large-scale genomic analysis places Chinese CC398 as a persistent human-associated MSSA lineage apart from the dominant global LA-MRSA clade.

Staphylococcus aureus clonal complex (CC)398 has emerged as a dominant livestock-associated methicillin-resistant S. aureus (LA-MRSA) lineage worldwide; however, its evolutionary trajectory and regional diversification remain incompletely understood. We developed a core-genome multilocus sequence typing (cgMLST) scheme with hierarchical clustering and applied it to over 30,000 S. aureus genomes, revealing frequent cross-border transmission of CC398. Subsequent time-calibrated phylogenetic analysis placed the most recent common ancestor at 1942 (95% CI: 1939-1945), with the human-to-livestock host jump around 1969 (95% CI: 1968-1972). Chinese CC398 exhibits a distinct trajectory: unlike the LA-MRSA lineages dominating Europe and North America, Chinese isolates are predominantly human-associated methicillin-susceptible S. aureus (HA-MSSA), forming unique East Asia-specific phylogroups (SAP1, SAP2, and AP1-AP3), with distinct resistance and virulence profiles. The LA lineage remains limited in China, with multinational mixed clusters emerging only after 2019. Analysis of global transmission networks revealed a significant correlation between LA-CC398 spread and international trade in fresh swine products, while no such correlation was observed for the human-associated lineage. Beyond the established lineage markers tet(M) and scn, our analysis identified additional differentially distributed genes, including cadC-a chromosomal cadmium resistance regulator-as a novel HA-lineage-enriched gene whose functional role in host adaptation remains to be determined. This study reveals that CC398 followed fundamentally different evolutionary paths in China versus Western countries, challenging a one-size-fits-all model of its dissemination.IMPORTANCEThis study illustrates how large-scale microbial genomics can resolve the evolutionary origins and regional diversification of bacterial pathogens. By applying a novel cgMLST scheme to over 30,000 S. aureus genomes, we show that CC398 followed fundamentally different evolutionary paths in China versus Western countries-challenging the prevailing model of uniform global dissemination-and that livestock-associated MRSA expansion is closely linked to international trade in fresh pork products. These findings highlight the need for integrated surveillance across human, animal, and trade interfaces to anticipate the emergence and spread of zoonotic pathogens.

Staphylococcus aureus↗
PubMed2026

Urine Proteomics as a Source of Biological Information and Outcome Predictor in Living Kidney Transplantation.

Kidney transplantation (KTx) is the preferred treatment for kidney failure. However, post-transplant management is challenging due to the limited lifespan of transplanted organs. Current methods for monitoring post-transplant complications are invasive and have limitations. Therefore, there is an urgent need for novel non-invasive biomarkers. This study investigates the proteomic composition of urine to understand renal biology during the process of transplantation and to identify potential markers for outcome prediction. Urine samples were collected from donors before transplantation and from recipients 4 weeks and 1 year after transplantation. Proteomic analysis was performed using mass spectrometry and label-free quantification. Statistical analyses included principal component analysis (PCA) and enrichment analysis. The resulting key findings were confirmed in an independent validation cohort. In addition, correlative regression models to evaluate the relationship between protein abundance and clinical outcomes in the further course after transplantation were performed. 106 urine samples in the setting of 70 kidney transplantations were analyzed. PCA revealed distinct clustering of donor and recipient samples, indicating significant proteomic changes after transplantation. Hierarchical clustering and gene ontology analysis identified molecular changes as a response to transplantation and showed an over-representation of relevant pathways related to inflammation, cell immune response and coagulation in both the original and validation cohorts. Multivariate regression analysis, including linear and logistic regression, identified 11 potential protein biomarkers, including ORM2, IL1RAP, APP, and FABP4 as predictors of eGFR 12 months after transplantation and 1 HP as a predictor of infections within the first year after transplantation, respectively. This study underscores the potential of non-invasive urine proteomics for identifying biological processes involved in kidney transplantation and for enhancing post-transplant monitoring and outcome prediction. We identified 12 potential biomarkers with added value to standard clinical parameters linked to transplant outcomes, which will be promising candidates for future outcome monitoring after KTx.

Humans↗
PubMed1978

Some personality patterns and dimensions of male alcoholics: a multivariate description.

The assumption that alcoholics form a homogeneous population has been found to be questionable. Recent research has been done to empirically define possible personality subtypes of alcoholics. This study extended the typological work done previously by Goldstein and Linden (1969a) and Whitelock, Overall, and Patrick (1971). They each found four alcoholic subtypes, three of which replicated across studies. For this research, MMPI profiles of 208 male alcoholics were submitted to a hierarchical clustering procedure. Seven subtypes were found. These results were compared to the results of the prior two studies, in addition to actuarial MMPI types previously delineated in clinical settings. Using a hierarchical factor analysis, these data were analyzed to determine the higher order interrelationships among MMPI scales for this alcoholic sample. These results were discussed, especially in terms of the implications for treatment and further research in alcoholism.

Adolescent↗
PubMed2026

Exploring cross-category relationships between symptoms in people with hypermobile EDS (hEDS) to identify disability patterns.

BACKGROUND: Hypermobile Ehlers-Danlos Syndrome (hEDS) is a connective tissue disorder with variable symptom presentation across multiple organ systems and significant morbidity. Little is known about hEDS etiology and identifying patterns of symptom co-occurrence can reveal previously unidentified relationships between phenotypes and inform studies of underlying disease pathophysiology for symptoms that may share functional biological pathways. In this exploratory analysis, we specifically assessed the distribution of symptoms in case and controls to identify clusters of co-occurring symptoms. METHODS: We have interrogated clinically relevant symptom areas in 47 females with hEDS, 36 age-matched female controls and 8 hypermobile patients without chronic pain. Studied symptoms include general health, mental health, body pain, vitality and energy, autonomic symptoms, bleeding, and gastrointestinal symptoms. We conducted hierarchal clustering on principle components (HCPC) to identify groups and compared the groups for the previously described symptoms. Radial plots were used to identify relationships between severe symptom categories. RESULTS: Our analysis reveals statistically significantly more severe symptoms in all categories in people with hEDS compared with age- and sex-matched controls and asymptomatic hypermobile patients. HCPC identified clearly separated Low, Moderate, and High symptom groups within participants. The Low dysfunction groups include nearly all controls and hypermobile patients without chronic pain. The High dysfunction group includes ~60% of people with hEDS, while around 40% are in the Moderate dysfunction cluster. Cluster solutions for all participants were stable with moderate fit (silhouette 0.64; Jaccard boot mean 0.91). Group level radial plots showed high bleeding severity across all symptom clusters, while disproportional severity of general health, physical function, limitation of role due to physical symptoms, pain, and social functioning deficits differentiates the High from Moderate and Low Dysfunction clusters. CONCLUSION: Using this analysis at the group level has revealed patterns suggesting a progression of disease symptoms. People with hypermobility do not uniformly have severe symptoms but instead have some symptoms that differentiate from non-hypermobile individuals. While exploratory, using a radar multi-symptom analysis may be used to evaluate disproportionately severe symptoms contributing to the patterns of global symptom severity. These include pain but also ability to perform roles, suggesting strong utility of physical and occupational therapies to emphasize coping. This may also allow better targeting of etiological studies and may have additional utility at an individual level to develop symptom management strategies.

Humans↗
PubMed2025

Host clustering of Campylobacter species and enteric pathogens in a longitudinal cohort of infants, family members and livestock in rural Eastern Ethiopia.

BACKGROUND: Livestock are recognized as major reservoirs for Campylobacter species and other enteric pathogens, posing infection risks to humans. High prevalence of Campylobacter during early childhood has been linked to environmental enteric dysfunction and stunting, particularly in low-resource settings. METHODS: A total of 280 samples from Campylobacter positive households with complete metadata were analyzed by shotgun metagenomic sequencing followed by bioinformatic analysis via the CZ-ID metagenomic pipeline (Illumina mNGS Pipeline v7.1). Further statistical analyses in JMP PRO 16 explored the microbiome, emphasizing Campylobacter and other enteric pathogens. Two-way hierarchical clustering and split k-mer analysis examined host structuring, patterns of co-infections and genetic relationships. Principal component analysis was used to characterize microbiome composition across the seven sample types. RESULTS: The study identified that microbiome composition was strongly host-driven, with more than 3844 genera detected, and two principal components explaining 62% of the total variation. Twenty-one dominant (based on relative abundance) Campylobacter species showed distinct clustering patterns for humans, ruminants, and broad hosts. The broad-host cluster included the most prevalent species, C. jejuni, C. concisus, and C. coli, present across sample types and a sub-cluster within C. jejuni involving humans, chickens, and ruminants. Campylobacter species from chickens showed strong positive correlations with mothers (r = 0.76), siblings (r = 0.61) and infants (r = 0.54), while co-occurrence analysis found a higher likelihood (Pr > 0.5) of pairs such as C. jejuni with C. coli, C. concisus, and C. showae. Analysis of the top 50 most abundant microbial taxa showed a distinct cluster uniquely present in human stool and absent in all livestock. The study also found frequent co-occurrence of C. jejuni with other enteric pathogens such as Salmonella, and Shigella, particularly in human and chicken. Additionally, instances of Candidatus Campylobacter infans (C. infans) were identified co-occurring with Salmonella and Shigella species in stool samples from infants, mothers, and siblings. CONCLUSIONS: A comprehensive analysis of Campylobacter diversity in humans and livestock in a low-resource setting revealed that infants can be exposed to multiple Campylobacter species early in life. C. jejuni is the dominant species with a propensity for co-occurrence with other notable enteric bacterial pathogens, including Salmonella, and Shigella, especially among infants. Video Abstract.

Animals↗
PubMed1978

Psychometrics of a neuropsychological test battery.

Compared factor analysis (linear) and hierarchical cluster analysis (nonlinear) of a neuropsychological battery of tests, including the Wechsler Adult Intelligence Scale, the Wechsler Memory Scale, the Graham-Kendall Memory for Designs Test, and the Bender Visual Motor Gestalt Designs Test. The results were discussed within the framework of descriptive and predictive analysis of the major cognitive functions of verbal intelligence, perceptual performance, and memory.

Bender-Gestalt Test↗
PubMed2026

Thrombus Metabolism-Based Molecular Subtyping for Prognostic Risk Stratification in Acute Ischemic Stroke: A Preliminary Study.

AIMS: To preliminarily characterize metabolic molecular subtypes of cerebral thromboemboli and evaluate their clinical significance in anterior circulation acute ischemic stroke due to large vessel occlusion (AIS-LVO). METHODS: Untargeted metabolomics was performed on thromboemboli retrieved from 36 patients with anterior circulation AIS-LVO using ultra-performance coupled liquid chromatography with quadrupole time-of-flight mass spectrometry (UPLC-Q-TOF-MS). Unsupervised hierarchical clustering was employed to identify distinct metabolic molecular subtypes, and their associations with stroke etiology, radiographic severity, and functional outcomes were analyzed. RESULTS: Two distinct thrombus metabolic molecular subtypes (C1 and C2) were identified based on 12 metabolites significantly associated with both short-term (7-day ∆NIHSS) and long-term (90-day mRS) functional outcomes. The C1 subtype, predominantly cardioembolic, exhibited enhanced lipid metabolism, whereas the C2 subtype, primarily atherothrombotic, demonstrated increased folate metabolism. Patients with C1 thromboemboli presented more severe admission ischemic lesions (as indicated by ASPECTS) and experienced poorer short-term and long-term outcomes. A six-metabolite signature derived from LASSO regression was identified for exploratory discrimination of thrombus metabolic subtypes, etiological subtypes, and 90-day outcomes. CONCLUSION: This preliminary exploratory study identifies two metabolically distinct thrombus molecular subtypes with clinical implications in anterior circulation AIS-LVO, providing a novel basis for risk stratification and personalized secondary prevention and warrants further investigation.

Humans↗
PubMed2025

Structural properties of short-chain carboxylic acids and alcohols relate to the molecular and physiological response of Salmonella enterica in an acidic environment.

Short-chain carboxylic acids (SCCA) and short-chain alcohols (SCALC) are naturally occurring antimicrobials that contribute to the biopreservation of food fermentations. This study investigated the effect of structurally different SCCA/SCALC with two-carbon (acetic acid; phenylacetic acid; 2-phenylethanol), three-carbon (propionic acid; 3-phenylpropionic acid; 3-phenylpropanol), and three-carbon chain with an additional hydroxyl group (lactic acid; 3-phenyllactic acid; 1-phenylpropanol) on the fitness, metabolic activity and gene expression of the pathogen Salmonella enterica at pH 4.5. SCCA inhibited Salmonella at lower concentrations than SCALC with the exception of lactic acid, which was partly consumed. The presence of a phenyl group enhanced antimicrobial activity. SCCA but not SCALC increased the lag phase of S. enterica, and in general, acetate was formed when cell growth was reduced by 20% suggesting a negative impact on bacteria fitness. Principal component analysis and hierarchical clustering indicated distinct gene expression profiles of S. enterica in response to SCCA or SCALC. In the presence of certain SCCA/SCALC, Salmonella activated pathways related to cellular pH control, and 1,2-propanediol, propionic acid and ethanolamine metabolism that involved the formation of metabolosomes. Genes related to flagellar assembly were less expressed and mobility was lower in the presence of lactic and 3-phenyllactic acid compared to controls suggesting a compound-specific response. KEY POINTS: • Differences in response among structurally different SCCA/SCALC at acidic condition. • SCCA/SCALC stress interfered with cell growth and metabolism of acetic and propionic acid. • Lactic acid prolonged the lag phase and reduced motility of Salmonella.

Salmonella enterica↗