Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Classification Algorithms”

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 523 records · Page 29Linked to original sources
PubMed2004

Comparing demographic, health status and psychosocial strategies of audience segmentation to promote physical activity.

The goal of audience segmentation is to identify population subgroups that are homogeneous with respect to certain variables associated with a given outcome or behavior. When such groups are identified and understood, targeted intervention strategies can be developed to address their unique characteristics and needs. This study compares the results of audience segmentation for physical activity that is based on either demographic, health status or psychosocial variables alone, or a combination of all three types of variables. Participants were 1090 African-American and White adults from two public health centers in St Louis, MO. Using a classification-tree algorithm to form homogeneous groups, analyses showed that more segments with greater variability in physical activity were created using psychosocial versus health status or demographic variables and that a combination of the three outperformed any individual set of variables. Simple segmentation strategies such as those relying on demographic variables alone provided little improvement over no segmentation at all. Audience segmentation appears to yield more homogeneous subgroups when psychosocial and health status factors are combined with demographic variables.

Adolescent↗
PubMed2006

An evaluation of contemporary hidden Markov model genefinders with a predicted exon taxonomy.

We present an independent evaluation of six recent hidden Markov model (HMM) genefinders. Each was tested on the new dataset (FSH298), the results of which showed no dramatic improvement over the genefinders tested five years ago. In addition, we introduce a comprehensive taxonomy of predicted exons and classify each resulting exon accordingly. These results are useful in measuring (with finer granularity) the effects of changes in a genefinder. We present an analysis of these results and identify four patterns of inaccuracy common in all HMM-based results.

Algorithms↗
PubMed2026

A sequence-based classifier distinguishes phenotype-associated genes from other gene models in plants.

Only a small fraction of annotated plant genes possess experimentally validated associations with specific phenotypes. Phenotype-associated genes have distinct structural, molecular, and evolutionary characteristics compared with nonvalidated gene models. Here, we develop a simple classifier that uses sequence and evolutionary features, which can be generated for any species with an annotated reference genome assembly, to accurately distinguish phenotype-associated genes from both the overall population of annotated gene models and a specific set of genes identified as being tolerant of premature stop mutations. A model trained solely on genes from maize (Zea mays) identifies and prioritizes rice (Oryza sativa) and Arabidopsis (Arabidopsis thaliana) genes that are highly enriched in genes with experimentally validated links to phenotypes in both of these evolutionarily distant species. Gene models predicted to have a higher probability of being linked to phenotypes display patterns consistent with known biological properties of phenotype-associated genes. Notably, the sets of genes predicted to have a high probability of being linked to phenotype variation do not consist exclusively of well-characterized gene families but included many uncharacterized gene families carrying domains of unknown function. The quantitative scores generated by this model offer a valuable resource for prioritizing and exploring the vast number of uncharacterized gene models in plants, reducing the risk of failure in future reverse genetic efforts and potentially accelerating gene discovery and functional annotation in crops.

Phenotype↗
PubMed1998

MEG covariance difference analysis: a method to extract target source activities by using task and control measurements.

A method is proposed for extracting target dipolesource activities from two sets of evoked magnetoencephalographic (MEG) data, one measured using task stimuli and the other using control stimuli. The difference matrix between the two covariance matrices obtained from these two measurements is calculated, and a procedure similar to the MEG-multiple signal classification (MUSIC) algorithm is applied to this difference matrix to extract the target dipole-source configuration. This configuration corresponds to the source-configuration difference between the two measurements. Computer simulation verified the validity of the proposed method. The method was applied to actual evoked-field data obtained from simulated task-and-control experiments. In these measurements, a combination of auditory and somatosensory stimuli was used as the task stimulus and the somatosensory stimulus alone was used as the control stimulus. The proposed covariance difference analysis successfully extracted the target auditory source and eliminated the disturbance from the somatosensory sources.

Algorithms↗
PubMed1998

Recursive MUSIC: a framework for EEG and MEG source localization.

The multiple signal classification (MUSIC) algorithm can be used to locate multiple asynchronous dipolar sources from electroencephalography (EEG) and magnetoencephalography (MEG) data. The algorithm scans a single-dipole model through a three-dimensional (3-D) head volume and computes projections onto an estimated signal subspace. To locate the sources, the user must search the head volume for multiple local peaks in the projection metric. This task is time consuming and subjective. Here, we describe an extension of this approach which we refer to as recursive MUSIC (R-MUSIC). This new procedure automatically extracts the locations of the sources through a recursive use of subspace projections. The new method is also able to locate synchronous sources through the use of a spatio-temporal independent topographies (IT) model. This model defines a source as one or more nonrotating dipoles with a single time course. Within this framework, we are able to locate fixed, rotating, and synchronous dipoles. The recursive subspace projection procedure that we introduce here uses the metric of canonical or subspace correlations as a multidimensional form of correlation analysis between the model subspace and the data subspace. By recursively computing subspace correlations, we build up a model for the sources which account for a given set of data. We demonstrate here how R-MUSIC can easily extract multiple asynchronous dipolar sources that are difficult to find using the original MUSIC scan. We then demonstrate R-MUSIC applied to the more general IT model and show results for combinations of fixed, rotating, and synchronous dipoles.

Algorithms↗
PubMed2003

Transmission mode time-reversal super-resolution imaging.

The theory of time-reversal super-resolution imaging of point targets embedded in a reciprocal background medium [A. J. Devaney, "Super-resolution imaging using time-reversal and MUSIC," J. Acoust. Soc. Am. (to be published)] is generalized to the case where the transmitter and receiver sensor arrays need not be coincident and for cases where the background medium can be nonreciprocal. The new theory developed herein is based on the singular value decomposition of the generalized multistatic data matrix of the sensor system rather than the standard eigenvector/eigenvalue decomposition of the time-reversal matrix as was employed in the above-mentioned work and other treatments of time-reversal imaging [Prada, Thomas, and Fink, "The iterative time reversal process: Analysis of the convergence," J. Acoust. Soc. Am. 97, 62 (1995); Prada et al., "Decomposition of the time reversal operator: Detection and selective focusing on two scatterers," J. Acoust. Soc. Am. 99, 2067 (1996)]. A generalized multiple signal classification (MUSIC) algorithm is derived that allows super-resolution imaging of both well-resolved and non-well-resolved point targets from arbitrary sensor array geometries. MUSIC exploits the orthogonal nature of the scatterer and noise subspaces defined by the singular vectors of the multistatic data matrix to form scatterer images. The time-reversal/MUSIC algorithm is tested and validated in two computer simulations of offset vertical seismic profiling where the sensor sources are aligned along the earth's surface and the receiver array is aligned along a subsurface borehole. All results demonstrate the high contrast, high resolution imaging capabilities of this new algorithm combination when compared with "classical" backpropagation or field focusing. Above and beyond the application of seismo-acoustic imaging, the time-reversal super-resolution theory has applications in ocean acoustics for target location, and ultrasonic nondestructive evaluation of parts.

Acoustics↗
PubMed2005

Introducing DOTUR, a computer program for defining operational taxonomic units and estimating species richness.

Although copious qualitative information describes the members of the diverse microbial communities on Earth, statistical approaches for quantifying and comparing the numbers and compositions of lineages in communities are lacking. We present a method that addresses the challenge of assigning sequences to operational taxonomic units (OTUs) based on the genetic distances between sequences. We developed a computer program, DOTUR, which assigns sequences to OTUs by using either the furthest, average, or nearest neighbor algorithm for each distance level. DOTUR uses the frequency at which each OTU is observed to construct rarefaction and collector's curves for various measures of richness and diversity. We analyzed 16S rRNA gene libraries derived from Scottish and Amazonian soils and the Sargasso Sea with DOTUR, which assigned sequences to OTUs rapidly and reliably based on the genetic distances between sequences and identified previous inconsistencies and errors in assigning sequences to OTUs. An analysis of the two 16S rRNA gene libraries from soil demonstrated that they do not contain enough sequences to support a claim that they contain different numbers of bacterial lineages with statistical confidence (P > 0.05), nor do they contain enough sequences to provide a robust estimate of species richness when an OTU is defined as containing sequences that are no more than 3% different from each other. In contrast, the richness of OTUs at the 3% level in the Sargasso Sea collection began to plateau after the sampling of 690 sequences. We anticipate that an equivalent extent of sampling for soil would require sampling more than 10,000 sequences, almost 100 times the size of typical sequence collections obtained from soil.

Algorithms↗
PubMed2004

Predicting spatial data with RBF networks.

Spatial prediction needs to account for spatial information, which makes conventional radial basis function (RBF) networks inappropriate, for they assume independent and identical distribution. In this paper, we fuse spatial information at different layers of RBF. Experiments show fusion at hidden layer gives the best result and suggest that the optimal value is around one for the coefficient, which is used in the linear combination at the output layer.

Algorithms↗
PubMed2026

Machine Learning and Metabolomics to Characterize Warburg-Like Metabolic Subtypes in Human Retinal Endothelial Cells Exposed to Risk Factors Associated With Proliferative Diabetic Retinopathy.

PURPOSE: High glucose (HG), hypoxia (Hyp), and their combination are major risk factors for proliferative diabetic retinopathy (PDR). Although these conditions induce features of the Warburg-like metabolic reprogramming in human retinal endothelial cells (HRECs), it remains unclear whether they produce distinct metabolic and angiogenic subtypes. This study aimed to characterize the Warburg-like-associated metabolic heterogeneity induced by these PDR-related risk factors and evaluate the ability of supervised machine-learning models to distinguish these subtypes. METHODS: HRECs were cultured under normoglycemic, HG, Hyp (2% O2), and combined HG-Hyp conditions. Untargeted LC-MS/MS metabolomics quantified metabolites spanning carbohydrates, amino acids, nucleotides, and lipids. Principal component analysis (PCA) assessed overall metabolic variation, and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis identified metabolic pathways associated with angiogenesis. In vitro angiogenesis assays measured endothelial tube formation and branching. Nine supervised classifiers (decision tree, logistic regression, naïve Bayes, random forest, K-Nearest Neighbors, neural network, gradient boosting, AdaBoost, and Support Vector Machine) were trained on the highest-ranked metabolites selected by the Information Gain Ratio feature-ranking approach. Model performance was evaluated using 10-fold cross-validation, leave-one-out cross-validation (LOOCV), permutation testing, and a classifier stability analysis under biologically meaningful distributional shift using an independent chemically induced hypoxia model (CoCl2). RESULTS: PCA revealed partial separation of metabolic profiles across conditions, indicating different Warburg-like metabolic subtypes. The combined HG-Hyp condition exhibited enhanced angiogenic potential relative to either HG or Hyp alone. KEGG pathway enrichment analysis identified fatty acid biosynthesis and elongation among the most significantly enriched pathways in HRECs under combined HG-Hyp conditions, alongside amino sugar and nucleotide sugar metabolism, glycerophospholipid metabolism, the pentose phosphate pathway, and glycolysis/gluconeogenesis. Supervised machine-learning classifiers distinguished these metabolic subtypes, with AdaBoost and gradient Boosting showing the most balanced, reproducible performance across 10-fold cross-validation, LOOCV, and permutation testing, and remaining the most reliable classifiers under domain-shift testing (area under the curve = 0.88, P = 0.0061). CONCLUSIONS: In this exploratory analysis, HG, Hyp, and their combination drive metabolically and functionally distinct subtypes of Warburg-like metabolic reprogramming in HRECs, with HG-Hyp in combination producing a highly angiogenic phenotype. Boosting-based ensemble classifiers provide a promising framework for detecting these subtypes even under domain-shift conditions, warranting validation in larger independent datasets. TRANSLATIONAL RELEVANCE: Integrating metabolomics with machine-learning classification offers a strategy to identify Warburg-like metabolic subtypes in retinal endothelial cells, providing insights into angiogenic mechanisms and guiding the development of targeted diagnostics or therapeutics for PDR.

Humans↗
PubMed2026

Machine learning-based clinical prediction model and multi-omics integration for assessing pancreatic cancer risk in new-onset diabetes.

BACKGROUND: Given that pancreatic cancer (PC) is typically diagnosed at an advanced stage but is often preceded by new-onset diabetes mellitus (NODM), providing a window for early detection, we sought to develop and validate an interpretable machine-learning model integrated with multi-omics profiling to identify early biomarkers of NODM-associated PC. METHODS: In a population-based cohort, individuals with NODM-associated PC and NODM without PC were identified and randomly divided (70:30) into training and validation sets after feature selection. Eight machine learning (ML) classifiers were compared using fivefold cross-validation, and model performance was evaluated in terms of discrimination, calibration, and decision curve–based clinical utility. We evaluated interpretability using the Shapley additive explanations (SHAP) analyses. Mechanistically, Olink proteomic profiling and metabolomics were analyzed through clinical classifications and model-defined risk strata. RESULTS: Categorical boosting achieved the best performance in the independent validation set (AUROC = 0.844). The NODM cohort was stratified into high- (n = 2,362) and low-risk (n = 5,030) groups, and internal validation together with SHAP analyses demonstrated consistent model performance and identified clinically interpretable predictors. Proteomic and metabolomic analyses under clinical and risk-based grouping identified 39 overlapping differentially expressed proteins and 145 overlapping metabolites with enriched across 11 shared KEGG pathways. Cross-platform validation highlighted PLTP, CRTAC1, and ITGAV as serum biomarkers with a strong potential for early NODM-PC detection. CONCLUSIONS: We developed an interpretable ML framework centered on NODM enables practical risk stratification for early PC detection by multi-omics and provides a pathway of ML-based triage followed by biomarker confirmation for earlier detection and diagnosis.

Humans↗
PubMed2026

A benchmarking study of feature screening approaches across type 1 diabetes omics studies classification settings.

In recent years, high dimensional omics analyses have become more commonplace for investigating complex biological systems. Typically, these studies attempt to identify key biomolecules associated with a particular biological process. Often, machine learning (ML) is used to identify these biomolecules, typically by learning which biomolecules are highly predictive of a treatment, biological outcome, or phenotype. A major challenge of applying ML to high throughput omics is overcoming noise when sample size is limited and unbalanced with respect to tens of thousands of biomolecules measured. Thus, feature selection (the process of reducing the number of predictors) is both a critical and common step in the ML analysis pipeline. While much attention has been given to embedding and wrapping techniques for feature selection in the omics space, filter-based methods for model-free feature selection have appealing theoretical properties. This manuscript evaluates sure screening, a class of filter-based feature selection methods which provide analytical guarantees for true feature set retention. Here, we cover existing feature screening methods based on the sure screening principal, available software, methods to improve feature screening, and contextualize feature screening in the larger discussion of feature selection for omics data analysis. Additionally, a suite of model-free sure screening approaches is applied and compared for several omics biomedical applications in a ML classification context. We identified BcorSIS as the most effective and computationally efficient screening method across various omics datasets, consistently outperforming others like CSIS and DCSIS in runtime.

Humans↗
PubMed2006

Medial profrontal cortex and anterior cingulate cortex in the generation of alpha activity induced by transcendental meditation: a magnetoencephalographic study.

Previous EEG studies have shown that transcendental meditation (TM) increases frontal and central alpha activity. The present study was aimed at identifying the source of this alpha activity using magnetoencephalography (MEG) and electroencephalography (EEG) simultaneously on eight TM practitioners before, during, and after TM. The magnetic field potentials corresponding to TM-induced alpha activities on EEG recordings were extracted, and we attempted to localize the dipole sources using the multiple signal classification (MUSIC) algorithm, equivalent current dipole source analysis, and the multiple spatio-temporal dipole model. Since the dipoles were mapped to both the medial prefrontal cortex (mPFC) and anterior cingulate cortex (ACC), it is suggested that the mPFC and ACC play an important role in brain activity induced by TM.

Adult↗
PubMed2004

Periportal fibrosis in human Schistosoma mansoni infection is associated with low IL-10, low IFN-gamma, high TNF-alpha, or low RANTES, depending on age and gender.

Schistosoma mansoni infection is highly endemic in parts of Uganda, and periportal fibrosis is common in communities along the shore of Lake Albert. In this study, we have identified cellular immune responses associated with fibrosis. A cohort of 199 individuals aged 6-50, resident in the village for at least 10 years or since birth, were examined for evidence of periportal fibrosis by ultrasound using the Niamey protocol. Whole-blood samples were assayed for levels of nine cellular immune molecules (IL-3, IL-4, IL-5, IL-10, IL-13, TNF-alpha, IFN-gamma, IL-1beta, and RANTES) in the absence of in vitro Ag stimulation, and after stimulation with egg and worm Ags. A lack of Ag specificity allowed the number of variables in the analysis to be reduced by factor analysis. The resulting factor scores were then entered into a risk analysis using a classification tree algorithm. Children, adult males, and adult females had different factors associated with fibrosis. Most cases of fibrosis in children (eight of nine) were associated with low (<47th percentile) IL-10 factor scores. Adult females at lowest risk had relatively high IFN-gamma factor scores (>83rd percentile), whereas those at highest risk had a combination of intermediate (32nd to 83rd percentile) IFN-gamma and relatively high (>60th percentile) TNF-alpha factor scores. Adult males at lowest risk of fibrosis had moderate TNF-alpha factor scores (55th to 82nd percentile), and a high risk was associated with either high TNF-alpha factor scores (>82nd percentile), or intermediate TNF-alpha combined with low RANTES factor scores (<58th percentile). These results demonstrate that periportal fibrosis is associated with cytokine production profiles that vary with both age and gender.

Adolescent↗
PubMed2001

The importance of time/space in diagnosing the causality of phylogenetic events: towards a "chronobiogeographical" paradigm?

A shift from a traditional biogeographical paradigm in cladistic biogeography to a chronobiogeographical paradigm is proposed. The chronobiogeographical paradigm aims to utilize temporal data in conjunction with spatial data in the detection of discrete historical events, such as vicariance and vicariant speciation, in cladograms. The concepts of primary and secondary congruency are introduced in relation to the distinction between repeated area relationships (primary congruency) and common extrinsic causality (secondary congruency). Simple hypothetical examples demonstrate that area cladograms cannot be safely interpreted purely as representing the sequence of area fragmentation; rather, they reflect recency of biotic interaction. Temporal data are shown to have a direct and potentially profound influence on the results of traditional cladistic biogeographical analyses, indicating the necessity of developing a chronobiogeographical approach. The implementation of the paradigm is considered first from a theoretical viewpoint and then in the context of the type of empirical data usually available. An as yet undevised "time/space algorithm" is deemed necessary for the latter, and guidelines are presented for the development of such an algorithm. Finally, we argue that the most rigorous and philosophically justified approach to the detection of phylogenetic causal events can be found only when temporal and spatial data are considered simultaneously. Consequently, the chronobiogeographical paradigm is seen as a logical elaboration of, not a replacement for, the biogeographical paradigm.

Algorithms↗
PubMedDate not supplied

Cecocolic torsion: classification, pathogenesis, and treatment.

Cecocolic torsion is a class of right colon obstruction. Under this heading, the known and the recently identified variants are unified under 1 classification. An algorithm is utilized to trace the pathogenesis of these variants. Recent data, prevailing definitions, and controversies are discussed and resolved. The significance of specific membranes and the blood supply is clarified. New symptoms, clinical and x-ray diagnoses are elucidated. The recommended treatments are outlined.

Algorithms↗
PubMed2005

How to distinguish parthood from location in bioontologies.

The pivotal role of the relation part-of in the description of living organisms is widely acknowledged. Organisms are open systems, which means that in contradistinction to mechanical artifacts they are characterized by a continuous flow and exchange of matter. A closer analysis of the spatial relations in biological organism reveals that the decision as to whether a given particular is part-of a second particular or whether it is only contained-in the second particular may often be controversial. We here propose a rule-based approach which allows us to decide on the basis of well-defined criteria which of the two relations holds between two anatomical objects, given that one spatially includes the other. We discuss the advantages and limitations of this approach, using concrete examples from human anatomy.

Algorithms↗
PubMed2006

AMIGOS: a method for the inspection of genomic organisation or structure and its application to characterise conserved gene arrangements.

In order to identify and to characterise gene clusters conserved in microbial genomes, the algorithm AMIGOS was developed. It is based on a categorisation of genes using a predefined set of gene functions (GFs). After the categorisation of all genes of a genome and based on their location on a replicon, distances between GFs were determined and stored in genome-specific matrices. These matrices were used to identify GF clusters like those strictly conserved in 13 archaeal, in 47 bacterial genomes and in the combination of the sets. Within the combined set of these 60 microbial genomes, there exist only two strictly conserved clusters harbouring two ribosomal genes each, namely those for L4, L23 and L22, L29. In order to characterise less strictly conserved GF clusters, content of genomes i.e. matrices were analysed pairwise. Resulting clusters were merged to (meta-) clusters if their content overlapped. A scoring system named cons(CL) was developed. It quantifies conservedness of cluster membership for individual GFs. For the genome of Escherichia coli it was shown that a grouping of cluster elements on cons(CL) values dissected the clusters into smaller sets. These sets were frequently overlapped by known transcriptional units (TUs). This finding justifies the usage of cons(CL) scores to predict TU membership of genes. In addition, cons(CL) values provide a sound basis for non-homologous gene annotation. Based on cons(CL) values, examples of conserved clusters containing annotated genes and single ones with unknown function are given.

Algorithms↗