SEARCH · PubMed Health

Results for “Clustering Algorithms”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

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

At least 1,459 records · Page 81Linked to original sources

PubMed1993

MASCOT: multiple alignment system for protein sequences based on three-way dynamic programming.

A multiple alignment methodology that can produce high-quality alignment is extremely important for predicting the structure of unknown proteins. Nearly all the methodologies developed so far have employed two-way alignment only. Although these methods are fast, the alignments they produce lose reliability as the similarity of sequences reduces. We developed the MASCOT multiple alignment system. MASCOT can sustain the reliability of alignment even when the similarity of sequences is low. MASCOT achieves high-quality alignment by employing three-way alignment in addition to two-way alignment. The resultant alignments are refined by simulated annealing to higher quality. We also use a cluster analysis of sequences to produce highly reliable alignments.

Algorithms↗

PubMed2004

Colloidal aggregation with sedimentation: concentration effects.

The results of computer models for colloidal aggregation, that consider both Brownian motion and gravitational drift experienced by the colloidal particles and clusters, are extended to include concentrations spanning three orders of magnitude. In previous publications and for a high colloidal concentration, it was obtained that the aggregation crosses over from diffusion-limited colloidal aggregation (DLCA) to another regime with a higher cluster fractal dimension and a speeding up followed by a slowing down of the aggregation rate. In the present work we show, as the concentration is decreased, that we can still cross over to a similar regime during the course of the aggregation, as long as the height of the sample is increased accordingly. Among the differences between the mentioned new regimes for a high and a low colloidal concentration, the cluster fractal dimension is higher for the high concentration case and lowers its value as the concentration is decreased, presumably reaching for low enough concentrations a fixed value above the DLCA value. It is also obtained the fractal dimension of the sediments, arising from the settling clusters that reach the bottom and continue a 2D-like diffusive motion and aggregation, on the floor of the container. For these clusters we now see two and sometimes three regimes, depending on concentration and sedimentation strength, with their corresponding fractal dimensions. The first two coming from the crossover already mentioned, that took place in the bulk of the sample before the cluster deposition, while the third arises from the two-dimensional aggregation on the floor of the container. For these bottom clusters we also obtain their dynamical behavior and aggregation rate.

Algorithms↗

PubMed2006

Ancient genomic architecture for mammalian olfactory receptor clusters.

BACKGROUND: Mammalian olfactory receptor (OR) genes reside in numerous genomic clusters of up to several dozen genes. Whole-genome sequence alignment nets of five mammals allow their comprehensive comparison, aimed at reconstructing the ancestral olfactory subgenome. RESULTS: We developed a new and general tool for genome-wide definition of genomic gene clusters conserved in multiple species. Syntenic orthologs, defined as gene pairs showing conservation of both genomic location and coding sequence, were subjected to a graph theory algorithm for discovering CLICs (clusters in conservation). When applied to ORs in five mammals, including the marsupial opossum, more than 90% of the OR genes were found within a framework of 48 multi-species CLICs, invoking a general conservation of gene order and composition. A detailed analysis of individual CLICs revealed multiple differences among species, interpretable through species-specific genomic rearrangements and reflecting complex mammalian evolutionary dynamics. One significant instance involves CLIC #1, which lacks a human member, implying the human-specific deletion of an OR cluster, whose mouse counterpart has been tentatively associated with isovaleric acid odorant detection. CONCLUSION: The identified multi-species CLICs demonstrate that most of the mammalian OR clusters have a common ancestry, preceding the split between marsupials and placental mammals. However, only two of these CLICs were capable of incorporating chicken OR genes, parsimoniously implying that all other CLICs emerged subsequent to the avian-mammalian divergence.

Animals↗

PubMed2003

Trajectory clustering: a non-parametric method for grouping gene expression time courses, with applications to mammary development.

Trajectory clustering is a novel and statistically well-founded method for clustering time series data from gene expression arrays. Trajectory clustering uses non-parametric statistics and is hence not sensitive to the particular distributions underlying gene expression data. Each cluster is clearly defined in terms of direction of change of expression for successive time points (its 'trajectory'), and therefore has easily appreciated biological meaning. Applying the method to a dataset from mouse mammary gland development, we demonstrate that it produces different clusters than Hierarchical, K-means, and Jackknife clustering methods, even when those methods are applied to differences between successive time points. Compared to all of the other methods, trajectory clustering was better able to match a manual clustering by a domain expert, and was better able to cluster groups of genes with known related functions.

Algorithms↗

PubMed2005

Medical record linkage in health information systems by approximate string matching and clustering.

BACKGROUND: Multiplication of data sources within heterogeneous healthcare information systems always results in redundant information, split among multiple databases. Our objective is to detect exact and approximate duplicates within identity records, in order to attain a better quality of information and to permit cross-linkage among stand-alone and clustered databases. Furthermore, we need to assist human decision making, by computing a value reflecting identity proximity. METHODS: The proposed method is in three steps. The first step is to standardise and to index elementary identity fields, using blocking variables, in order to speed up information analysis. The second is to match similar pair records, relying on a global similarity value taken from the Porter-Jaro-Winkler algorithm. And the third is to create clusters of coherent related records, using graph drawing, agglomerative clustering methods and partitioning methods. RESULTS: The batch analysis of 300,000 "supposedly" distinct identities isolates 240,000 true unique records, 24,000 duplicates (clusters composed of 2 records) and 3,000 clusters whose size is greater than or equal to 3 records. CONCLUSION: Duplicate-free databases, used in conjunction with relevant indexes and similarity values, allow immediate (i.e. real-time) proximity detection when inserting a new identity.

Algorithms↗

PubMed2005

Population-based continuous optimization, probabilistic modelling and mean shift.

Evolutionary algorithms perform optimization using a population of sample solution points. An interesting development has been to view population-based optimization as the process of evolving an explicit, probabilistic model of the search space. This paper investigates a formal basis for continuous, population-based optimization in terms of a stochastic gradient descent on the Kullback-Leibler divergence between the model probability density and the objective function, represented as an unknown density of assumed form. This leads to an update rule that is related and compared with previous theoretical work, a continuous version of the population-based incremental learning algorithm, and the generalized mean shift clustering framework. Experimental results are presented that demonstrate the dynamics of the new algorithm on a set of simple test problems.

Algorithms↗

PubMed2000

Fuzzy EMG classification for prosthesis control.

This paper proposes a fuzzy approach to classify single-site electromyograph (EMG) signals for multifunctional prosthesis control. While the classification problem is the focus of this paper, the ultimate goal is to improve myoelectric system control performance, and classification is an essential step in the control. Time segmented features are fed to a fuzzy system for training and classification. In order to obtain acceptable training speed and realistic fuzzy system structure, these features are clustered without supervision using the Basic Isodata algorithm at the beginning of the training phase, and the clustering results are used in initializing the fuzzy system parameters. Afterwards, fuzzy rules in the system are trained with the back-propagation algorithm. The fuzzy approach was compared with an artificial neural network (ANN) method on four subjects, and very similar classification results were obtained. It is superior to the latter in at least three points: slightly higher recognition rate; insensitivity to overtraining; and consistent outputs demonstrating higher reliability. Some potential advantages of the fuzzy approach over the ANN approach are also discussed.

Algorithms↗

PubMedDate not supplied

Combining sequence and time series expression data to learn transcriptional modules.

Our goal is to cluster genes into transcriptional modules--sets of genes where similarity in expression is explained by common regulatory mechanisms at the transcriptional level. We want to learn modules from both time series gene expression data and genome-wide motif data that are now readily available for organisms such as S. cereviseae as a result of prior computational studies or experimental results. We present a generative probabilistic model for combining regulatory sequence and time series expression data to cluster genes into coherent transcriptional modules. Starting with a set of motifs representing known or putative regulatory elements (transcription factor binding sites) and the counts of occurrences of these motifs in each gene's promoter region, together with a time series expression profile for each gene, the learning algorithm uses expectation maximization to learn module assignments based on both types of data. We also present a technique based on the Jensen-Shannon entropy contributions of motifs in the learned model for associating the most significant motifs to each module. Thus, the algorithm gives a global approach for associating sets of regulatory elements to "modules" of genes with similar time series expression profiles. The model for expression data exploits our prior belief of smooth dependence on time by using statistical splines and is suitable for typical time course data sets with relatively few experiments. Moreover, the model is sufficiently interpretable that we can understand how both sequence data and expression data contribute to the cluster assignments, and how to interpolate between the two data sources. We present experimental results on the yeast cell cycle to validate our method and find that our combined expression and motif clustering algorithm discovers modules with both coherent expression and similar motif patterns, including binding motifs associated to known cell cycle transcription factors.

Algorithms↗

PubMed2004

Neural spike classification using parallel selection of all algorithm parameters.

The Forster-Handwerker template-matching algorithm (J. Neurosci. Methods 31 (1990) 109) classifies neuronal spikes according to three parameters selected by the experimenter prior to running the algorithm. Thousands of different combinations of these parameter values are possible producing hundreds of different classifications for each input file. Using a 40-processor Linux-based parallel computing cluster, we ran their algorithm with an effective sampling of all combinations of parameter values in order to generate a list of the classifications that can be generated by the algorithm. A distance measure was used to quantify the similarity between classifications and then to create a distance table containing entries for the distances between all pairs of classifications. Using a self-organizing neural network (SON) and the distance table we group the classifications by similarity and select the best representative classifications that the Forster-Handwerker algorithm can produce.

Action Potentials↗

PubMed2005

Empirical modeling of human face kinematics during speech using motion clustering.

In this paper we present an algorithm for building an empirical model of facial biomechanics from a set of displacement records of markers located on the face of a subject producing speech. Markers are grouped into clusters, which have a unique primary marker and a number of secondary markers with an associated weight. Motion of the secondary markers is computed as the weighted sum of the primary markers of the clusters to which they belong. This model may be used to produce facial animations, by driving the primary markers with measured kinematic signals.

Algorithms↗

PubMed2006

A descriptive method to evaluate the number of regimes in a switching autoregressive model.

This paper proposes a descriptive method for an open problem in time series analysis: determining the number of regimes in a switching autoregressive model. We will translate this problem into a classification one and define a criterion for hierarchically clustering different model fittings. Finally, the method will be tested on simulated examples and real-life data.

Algorithms↗

PubMed2006

STEM: a tool for the analysis of short time series gene expression data.

BACKGROUND: Time series microarray experiments are widely used to study dynamical biological processes. Due to the cost of microarray experiments, and also in some cases the limited availability of biological material, about 80% of microarray time series experiments are short (3-8 time points). Previously short time series gene expression data has been mainly analyzed using more general gene expression analysis tools not designed for the unique challenges and opportunities inherent in short time series gene expression data. RESULTS: We introduce the Short Time-series Expression Miner (STEM) the first software program specifically designed for the analysis of short time series microarray gene expression data. STEM implements unique methods to cluster, compare, and visualize such data. STEM also supports efficient and statistically rigorous biological interpretations of short time series data through its integration with the Gene Ontology. CONCLUSION: The unique algorithms STEM implements to cluster and compare short time series gene expression data combined with its visualization capabilities and integration with the Gene Ontology should make STEM useful in the analysis of data from a significant portion of all microarray studies. STEM is available for download for free to academic and non-profit users at http://www.cs.cmu.edu/~jernst/stem.

Cluster Analysis↗

PubMed2006

SPATCLUS: an R package for arbitrarily shaped multiple spatial cluster detection for case event data.

This paper describes an R package, named SPATCLUS that implements a method recently proposed for spatial cluster detection of case event data. This method is based on a data transformation. This transformation is achieved by the definition of a trajectory, which allows to attribute to each point a selection order and the distance to its nearest neighbour. The nearest point is searched among the points which have not yet been selected in the trajectory. Due to the trajectory effects, the distance is weighted by the expected distance under the uniform distribution hypothesis. Potential clusters are located by using multiple structural change models and a dynamic programming algorithm. The double maximum test allows to select the best model. The significativity of potential clusters is determined by Monte Carlo simulations. This method makes it possible the detection of multiple clusters of any shape.

Algorithms↗

PubMed1999

Unsupervised visual learning of three-dimensional objects using a modular network architecture.

This paper presents a modular network architecture that learns to cluster multiple views of multiple three-dimensional (3D) objects. The proposed network model is based on a mixture of non-linear autoencoders, which compete to encode multiple views of each 3D object. The main advantage of using a mixture of autoencoders is that it can capture multiple non-linear sub-spaces, rather than multiple centers for describing complex shapes of the view distributions. The unsupervised training algorithm is formulated within a maximum-likelihood estimation framework. The performance of the modular network model is evaluated through experiments using synthetic 3D wire-frame objects and gray-level images of real 3D objects. It is shown that the performance of the modular network model is superior to the performance of the conventional clustering algorithms, such as the K-means algorithm and the Gaussian mixture model.

Journal Article↗

PubMed2000

Improvements to the sensitivity of gravitational clustering for multiple neuron recordings.

We outline two improvements to the technique of gravitational clustering for detection of neuronal synchrony, which are capable of improving the method's detection of weak synchrony with limited data. The advantages of the enhancements are illustrated using data with known levels of synchrony and different interspike interval distributions. The novel simulation method described can easily generate such test data. An important dependence of the sensitivity of gravitational clustering to the interspike interval distribution of the analysed spike trains is described.

Action Potentials↗

PubMed1998

Quantitative characterization, classification and reconstruction of oocyst shapes of Eimeria species from cattle.

This study reports on morphological variability of Eimeria species, which may be given either by drawings or as quantitative data. The drawings may be used to facilitate identification by eye of 'unknown' Eimeria specimens, whereas quantitative data may serve as a reference set for identification by multivariate statistical techniques. The morphology of 810 Eimeria specimens was defined in binary (b/w) digital images by pixels of their oocyst outline. A Fourier transform of pixel positions yielded size and shape features. To classify coccidia, the quantitative data were employed in an agglomerative clustering by average linkage algorithm with equal weight assigned to size and shape. An inverse Fourier transform served to reconstruct oocyst outlines, i.e. outlines of average shape and size, from mean values of features in resulting clusters. Clusters were subsequently identified based on their average morphology by comparison with drawings of species in an earlier taxonomical work. Five hundred oocyst outlines were simulated for each cluster representing a species, and shape/size variability was presented in contour diagrams. Differences in species shapes, and correspondence in length and width, were seen after reconstruction by inverse Fourier transform and comparison with earlier studies.

Animals↗

PubMed2004

Structural and computational depth of diffusion-limited aggregation.

Diffusion-limited aggregation (DLA) is studied from the perspective of computational complexity. A parallel algorithm is exhibited that requires a number of steps that scales as the depth of the tree defined by the cluster. The existence of this algorithm suggests a connection between a fundamental computational and structural property of DLA.

Algorithms↗

PubMedDate not supplied

Attribute clustering for grouping, selection, and classification of gene expression data.

This paper presents an attribute clustering method which is able to group genes based on their interdependence so as to mine meaningful patterns from the gene expression data. It can be used for gene grouping, selection, and classification. The partitioning of a relational table into attribute subgroups allows a small number of attributes within or across the groups to be selected for analysis. By clustering attributes, the search dimension of a data mining algorithm is reduced. The reduction of search dimension is especially important to data mining in gene expression data because such data typically consist of a huge number of genes (attributes) and a small number of gene expression profiles (tuples). Most data mining algorithms are typically developed and optimized to scale to the number of tuples instead of the number of attributes. The situation becomes even worse when the number of attributes overwhelms the number of tuples, in which case, the likelihood of reporting patterns that are actually irrelevant due to chances becomes rather high. It is for the aforementioned reasons that gene grouping and selection are important preprocessing steps for many data mining algorithms to be effective when applied to gene expression data. This paper defines the problem of attribute clustering and introduces a methodology to solving it. Our proposed method groups interdependent attributes into clusters by optimizing a criterion function derived from an information measure that reflects the interdependence between attributes. By applying our algorithm to gene expression data, meaningful clusters of genes are discovered. The grouping of genes based on attribute interdependence within group helps to capture different aspects of gene association patterns in each group. Significant genes selected from each group then contain useful information for gene expression classification and identification. To evaluate the performance of the proposed approach, we applied it to two well-known gene expression data sets and compared our results with those obtained by other methods. Our experiments show that the proposed method is able to find the meaningful clusters of genes. By selecting a subset of genes which have high multiple-interdependence with others within clusters, significant classification information can be obtained. Thus, a small pool of selected genes can be used to build classifiers with very high classification rate. From the pool, gene expressions of different categories can be identified.

Algorithms↗