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 1,207 records · Page 67Linked to original sources

PubMed2006

Mining gene expression data by interpreting principal components.

BACKGROUND: There are many methods for analyzing microarray data that group together genes having similar patterns of expression over all conditions tested. However, in many instances the biologically important goal is to identify relatively small sets of genes that share coherent expression across only some conditions, rather than all or most conditions as required in traditional clustering; e.g. genes that are highly up-regulated and/or down-regulated similarly across only a subset of conditions. Equally important is the need to learn which conditions are the decisive ones in forming such gene sets of interest, and how they relate to diverse conditional covariates, such as disease diagnosis or prognosis. RESULTS: We present a method for automatically identifying such candidate sets of biologically relevant genes using a combination of principal components analysis and information theoretic metrics. To enable easy use of our methods, we have developed a data analysis package that facilitates visualization and subsequent data mining of the independent sources of significant variation present in gene microarray expression datasets (or in any other similarly structured high-dimensional dataset). We applied these tools to two public datasets, and highlight sets of genes most affected by specific subsets of conditions (e.g. tissues, treatments, samples, etc.). Statistically significant associations for highlighted gene sets were shown via global analysis for Gene Ontology term enrichment. Together with covariate associations, the tool provides a basis for building testable hypotheses about the biological or experimental causes of observed variation. CONCLUSION: We provide an unsupervised data mining technique for diverse microarray expression datasets that is distinct from major methods now in routine use. In test uses, this method, based on publicly available gene annotations, appears to identify numerous sets of biologically relevant genes. It has proven especially valuable in instances where there are many diverse conditions (10's to hundreds of different tissues or cell types), a situation in which many clustering and ordering algorithms become problematic. This approach also shows promise in other topic domains such as multi-spectral imaging datasets.

Algorithms↗

PubMed2006

Gene selection and classification of microarray data using random forest.

BACKGROUND: Selection of relevant genes for sample classification is a common task in most gene expression studies, where researchers try to identify the smallest possible set of genes that can still achieve good predictive performance (for instance, for future use with diagnostic purposes in clinical practice). Many gene selection approaches use univariate (gene-by-gene) rankings of gene relevance and arbitrary thresholds to select the number of genes, can only be applied to two-class problems, and use gene selection ranking criteria unrelated to the classification algorithm. In contrast, random forest is a classification algorithm well suited for microarray data: it shows excellent performance even when most predictive variables are noise, can be used when the number of variables is much larger than the number of observations and in problems involving more than two classes, and returns measures of variable importance. Thus, it is important to understand the performance of random forest with microarray data and its possible use for gene selection. RESULTS: We investigate the use of random forest for classification of microarray data (including multi-class problems) and propose a new method of gene selection in classification problems based on random forest. Using simulated and nine microarray data sets we show that random forest has comparable performance to other classification methods, including DLDA, KNN, and SVM, and that the new gene selection procedure yields very small sets of genes (often smaller than alternative methods) while preserving predictive accuracy. CONCLUSION: Because of its performance and features, random forest and gene selection using random forest should probably become part of the "standard tool-box" of methods for class prediction and gene selection with microarray data.

Algorithms↗

PubMed2006

Probe-level linear model fitting and mixture modeling results in high accuracy detection of differential gene expression.

BACKGROUND: The identification of differentially expressed genes (DEGs) from Affymetrix GeneChips arrays is currently done by first computing expression levels from the low-level probe intensities, then deriving significance by comparing these expression levels between conditions. The proposed PL-LM (Probe-Level Linear Model) method implements a linear model applied on the probe-level data to directly estimate the treatment effect. A finite mixture of Gaussian components is then used to identify DEGs using the coefficients estimated by the linear model. This approach can readily be applied to experimental design with or without replication. RESULTS: On a wholly defined dataset, the PL-LM method was able to identify 75% of the differentially expressed genes within 10% of false positives. This accuracy was achieved both using the three replicates per conditions available in the dataset and using only one replicate per condition. CONCLUSION: The method achieves, on this dataset, a higher accuracy than the best set of tools identified by the authors of the dataset, and does so using only one replicate per condition.

Algorithms↗

PubMed2006

Semi-supervised discovery of differential genes.

BACKGROUND: Various statistical scores have been proposed for evaluating the significance of genes that may exhibit differential expression between two or more controlled conditions. However, in many clinical studies to detect clinical marker genes for example, the conditions have not necessarily been controlled well, thus condition labels are sometimes hard to obtain due to physical, financial, and time costs. In such a situation, we can consider an unsupervised case where labels are not available or a semi-supervised case where labels are available for a part of the whole sample set, rather than a well-studied supervised case where all samples have their labels. RESULTS: We assume a latent variable model for the expression of active genes and apply the optimal discovery procedure (ODP) proposed by Storey (2005) to the model. Our latent variable model allows gene significance scores to be applied to unsupervised and semi-supervised cases. The ODP framework improves detectability by sharing the estimated parameters of null and alternative models of multiple tests over multiple genes. A theoretical consideration leads to two different interpretations of the latent variable, i.e., it only implicitly affects the alternative model through the model parameters, or it is explicitly included in the alternative model, so that the interpretations correspond to two different implementations of ODP. By comparing the two implementations through experiments with simulation data, we have found that sharing the latent variable estimation is effective for increasing the detectability of truly active genes. We also show that the unsupervised and semi-supervised rating of genes, which takes into account the samples without condition labels, can improve detection of active genes in real gene discovery problems. CONCLUSION: The experimental results indicate that the ODP framework is effective for hypotheses including latent variables and is further improved by sharing the estimations of hidden variables over multiple tests.

Algorithms↗

PubMed2005

Computational tradeoffs in multiplex PCR assay design for SNP genotyping.

BACKGROUND: Multiplex PCR is a key technology for detecting infectious microorganisms, whole-genome sequencing, forensic analysis, and for enabling flexible yet low-cost genotyping. However, the design of a multiplex PCR assays requires the consideration of multiple competing objectives and physical constraints, and extensive computational analysis must be performed in order to identify the possible formation of primer-dimers that can negatively impact product yield. RESULTS: This paper examines the computational design limits of multiplex PCR in the context of SNP genotyping and examines tradeoffs associated with several key design factors including multiplexing level (the number of primer pairs per tube), coverage (the % of SNP whose associated primers are actually assigned to one of several available tube), and tube-size uniformity. We also examine how design performance depends on the total number of available SNPs from which to choose, and primer stringency criterial. We show that finding high-multiplexing/high-coverage designs is subject to a computational phase transition, becoming dramatically more difficult when the probability of primer pair interaction exceeds a critical threshold. The precise location of this critical transition point depends on the number of available SNPs and the level of multiplexing required. We also demonstrate how coverage performance is impacted by the number of available snps, primer selection criteria, and target multiplexing levels. CONCLUSION: The presence of a phase transition suggests limits to scaling Multiplex PCR performance for high-throughput genomics applications. Achieving broad SNP coverage rapidly transitions from being very easy to very hard as the target multiplexing level (# of primer pairs per tube) increases. The onset of a phase transition can be "delayed" by having a larger pool of SNPs, or loosening primer selection constraints so as to increase the number of candidate primer pairs per SNP, though the latter may produce other adverse effects. The resulting design performance tradeoffs define a benchmark that can serve as the basis for comparing competing multiplex PCR design optimization algorithms and can also provide general rules-of-thumb to experimentalists seeking to understand the performance limits of standard multiplex PCR.

Algorithms↗

PubMed2005

A generic approach for the design of whole-genome oligoarrays, validated for genomotyping, deletion mapping and gene expression analysis on Staphylococcus aureus.

BACKGROUND: DNA microarray technology is widely used to determine the expression levels of thousands of genes in a single experiment, for a broad range of organisms. Optimal design of immobilized nucleic acids has a direct impact on the reliability of microarray results. However, despite small genome size and complexity, prokaryotic organisms are not frequently studied to validate selected bioinformatics approaches. Relying on parameters shown to affect the hybridization of nucleic acids, we designed freely available software and validated experimentally its performance on the bacterial pathogen Staphylococcus aureus. RESULTS: We describe an efficient procedure for selecting 40-60 mer oligonucleotide probes combining optimal thermodynamic properties with high target specificity, suitable for genomic studies of microbial species. The algorithm for filtering probes from extensive oligonucleotides libraries fitting standard thermodynamic criteria includes positional information of predicted target-probe binding regions. This algorithm efficiently selected probes recognizing homologous gene targets across three different sequenced genomes of Staphylococcus aureus. BLAST analysis of the final selection of 5,427 probes yielded >97%, 93%, and 81% of Staphylococcus aureus genome coverage in strains N315, Mu50, and COL, respectively. A manufactured oligoarray including a subset of control Escherichia coli probes was validated for applications in the fields of comparative genomics and molecular epidemiology, mapping of deletion mutations and transcription profiling. CONCLUSION: This generic chip-design process merging sequence information from several related genomes improves genome coverage even in conserved regions.

Algorithms↗

PubMed2001

A computationally efficient method for obtaining standard error estimates for the promax and related solutions.

A computationally efficient algorithm for computing the asymptotic standard errors for the promax factor solution is proposed. The algorithm covers promax rotation with or without row normalization of the pre-rotated factor matrix. It also covers situations with either even- or odd-powered promax targets. With some modifications, the algorithm applies to Procrustean rotations with fixed or random independent targets. Simulation results show that the algorithm provides reasonable approximate standard errors for the promax solution with N = 200. In a real data example, the numerical results of the standard error computation using the proposed algorithm match those of an existing method based largely on the augmented information approach. The reasons why the proposed algorithm is more efficient than the augmented information approach are discussed.

Algorithms↗

PubMed2002

Three-dimensional-line skeleton graph analysis of high-resolution magnetic resonance images: a validation study from 34-microm-resolution microcomputed tomography.

The resolution achievable in vivo by magnetic resonance imaging (MRI) techniques is not sufficient to depict precisely individual trabeculae and, thus, does not permit the quantification of the "true" trabecular bone morphology and topology. Nevertheless, the characterization of the "apparent" trabecular bone network derived from high-resolution MR images (MRIs) and their potential to provide information in addition to bone mineral density (BMD) alone has been established in studies of osteoporosis. The aim of this work was to show the ability of the three-dimensional-line skeleton graph analysis (3D-LSGA) to characterize high-resolution MRIs of trabecular bone structure. Fifteen trabecular bone samples of the distal radius were imaged using the high-resolution MRI (156 x 156 x 300 microm3) and microcomputed tomography (microCT; 34 x 34 x 34 microm3). After thresholding, the 3D skeleton graph of each binary image was obtained. To remove the assimilated-noise branches of the skeleton graph and smooth this skeleton graph before it was analyzed, we defined a smoothing length criterion (l(c)), such that all "termini" branches having a length lower than l(c) were removed. Local topological and morphological LSGA measurements were performed from MRIs and microCT images of the same samples. The correlations between these two sets of measurements were dependent on the smoothing criterion l(c), reaching R2 = 0.85 for topological measurements and R2 = 0.57-0.64 for morphological measurements. 3D-LSGA technique could be applied to in vivo high-resolution MRIs of trabecular bone structure, giving an indirect characterization of the microtrabecular bone network.

Adult↗

PubMed2004

Parameter-optimized digital holographic microscope for high-resolution living-cell analysis.

A parameter-optimized off-axis setup for digital holographic microscopy is presented for simultaneous, high-resolution, full-field quantitative amplitude and quantitative phase-contrast microscopy and the detection of changes in optical path length in transparent objects, such as undyed living cells. Numerical reconstruction with the described nondiffractive reconstruction method, which suppresses the zero order and the twin image, requires a mathematical model of the phase-difference distribution between the object wave and the reference wave in the hologram plane. Therefore an automated algorithm is explained that determines the parameters of the mathematical model by carrying out the discrete Fresnel transform. Furthermore the relationship between the axial position of the object and the reconstruction distance, which is required for optimization of the lateral resolution of the holographic images, is derived. The lateral and the axial resolutions of the system are discussed and quantified by application to technical objects and to living cells.

Algorithms↗

PubMed2005

Spectrum analysis of motion parallax in a 3D cluttered scene and application to egomotion.

Previous methods for estimating observer motion in a rigid 3D scene assume that image velocities can be measured at isolated points. When the observer is moving through a cluttered 3D scene such as a forest, however, pointwise measurements of image velocity are more challenging to obtain because multiple depths, and hence multiple velocities, are present in most local image regions. We introduce a method for estimating egomotion that avoids pointwise image velocity estimation as a first step. In its place, the direction of motion parallax in local image regions is estimated, using a spectrum-based method, and these directions are then combined to directly estimate 3D observer motion. There are two advantages to this approach. First, the method can be applied to a wide range of 3D cluttered scenes, including those for which pointwise image velocities cannot be measured because only normal velocity information is available. Second, the egomotion estimates can be used as a posterior constraint on estimating pointwise image velocities, since known egomotion parameters constrain the candidate image velocities at each point to a one-dimensional rather than a two-dimensional space.

Algorithms↗

PubMed2003

New method for pollen identification by FT-IR spectroscopy.

A new methodology for identification of pollen was developed based on FT-IR spectroscopy. Pollen samples of twenty different plant species were collected and the diffuse reflectance infrared Fourier transform (DRIFTS) and KBr pellet spectra were recorded. Libraries of spectra were created. Spectra of unknown plant origin pollen were recorded and compared with those of the corresponding pollen library and the match value was measured automatically using the appropriate software (OMINC ver. 3.1). From the same pollen samples, microscopic slides were prepared and the photographs of the pollen grains were used as a second comparison method. Using light microscopy, the pollen identification is usually limited to the family or generic name, while FT-IR spectroscopy can distinguish species belonging to the same genus. This method is simple and fast, and when the DRIFTS technique is used the sample is not destroyed.

Algorithms↗

PubMed2006

Differentiation of selected Salmonella enterica serovars by Fourier transform mid-infrared spectroscopy.

Salmonella enterica serovars include pathogens responsible for high numbers of foodborne salmonellosis. Fourier transform infrared (FT-IR) spectroscopy can be used to rapidly and accurately identify microorganisms based on unique spectra of bacterial cell components. The objectives of this study were to discriminate closely related Salmonella enterica serovars by using FT-IR spectroscopy and multivariate analysis and to compare the performance of three techniques for differentiating among Salmonella serovars. Selected serovars of S. enterica were streaked onto plate count agar and incubated (37 degrees C, 24 h). Isolated colonies were suspended in phosphate buffer or 50% ethanol (10 microL). Suspensions were placed on (1) ZnSe crystals for transmission, (2) disposable polyethylene membranes (DPM) for transmission, and (3) diamond crystal plate for attenuated total reflectance (ATR) analyses; all samples were dried under vacuum. Classification models, soft independent modeling of class analogy (SIMCA), from derivatized infrared spectra (1300-900 cm(-1)), discriminated among Salmonella serovars presumably attributed to cell's lipopolysaccharides (1000-980 cm(-1)). Samples on DPM required high cell density for reliable spectra. High-quality spectra were obtained when a single colony was suspended in ethanol or buffer and mounted on ZnSe crystals for transmission or diamond plate for ATR analysis. Prediction of unknowns, representative of serovars used to construct classification models, showed that all techniques were suitable for the rapid and accurate differentiation of Salmonella serovars.

Algorithms↗

PubMed2026

Motif-Cluster: Motif driven prioritization of transcription factor binding clusters.

Genome-wide analyses of transcription factor (TF) motif binding sites have largely emphasized individual high-affinity sites, while overlooking the regulatory importance of locally repetitive motif clusters. Such clusters, including combinations of weak and strong binding sites, can collectively enhance TF occupancy and regulatory activity. Here we present Motif-Cluster, an open-source framework for motif-driven prioritization and visualization of TF binding clusters using sequence information alone. Motif-Cluster integrates a density-based clustering strategy with flexible modeling of binding-site gaps and affinity signals, enabling the identification and ranking of candidate regulatory regions without requiring experimental binding data. Through simulations and multiple real-data analyses, we show that combining gap distributions with binding affinity effectively balances cluster size and signal strength while reducing noise from weak sites. Application to ZNF410 successfully recovers the previously characterized binding clusters in the CHD4 promoter, which are conserved between human and mouse. Additional case studies involving PHB1, TWIST1, and EGR1 further demonstrate the general applicability of the method across diverse transcription factors. Motif-Cluster also provides intuitive visualization and reproducible workflows to facilitate interpretation of spatially dense motif patterns. Overall, Motif-Cluster offers a robust and flexible approach for prioritizing transcription factor regulatory regions from genome-wide motif scans, enabling biological discovery and guiding experimental design, particularly in settings where direct genome-wide binding assays are unavailable.

Transcription Factors↗

PubMed1991

LH pulsatility and in vitro bioactivity in women with anorexia nervosa-related hypothalamic amenorrhea.

LH nocturnal pulsatility and bioactivity to immunoreactivity (B/I) ratio were determined in 16 patients with anorexia nervosa-related hypothalamic amenorrhea and low sex steroid levels, and in 12 normal women in the midfollicular phase. The patients were subdivided into 2 groups: IA (N = 7) without, and IB (N = 9) with documented recent weight gain. Blood samples were taken from each subject at 10-min intervals from 00.00 to 06.00 h. Immunoreactive LH data were analysed with cluster analysis algorithm. A pool of aliquots from all the samples was used to evaluate bioactive LH, immunoreactive LH and LH B/I ratio in each subject. LH pulse frequency was lower in Group IA than in controls, whereas it did not differ significantly between Group IB and controls. LH pulse amplitude was lower in Group IA, and higher in Group IB than in controls. LH B/I ratio was below the control range in 3/16 patients. In conclusion, persistent hypothalamic amenorrhea does not require a permanent inhibition of the GnRH pulse generator; transient inhibition of pulsatility and qualitative abnormalities of gonadotropins could be involved in the mechanism, at least in some patients.

Adolescent↗

PubMed1992

Role of ovarian sex steroids in the regulation of thyrotropin (TSH) secretion of hypogonadal women.

The exact role of ovarian sex steroids in the neuroendocrine regulation of thyrotropin (TSH) release in women can only be accurately assessed in the absence of any considerable ovarian sex steroid feedback upon the hypothalamic-pituitary unit. Consequently, the unstimulated episodic and thyrotropin-releasing hormone (TRH) stimulated TSH secretion was evaluated in postmenopausal women before and during sequential ovarian sex steroid replacements. Seven euthyroid women (mean age: 59.4 years) were studied initially without any sex steroid replacement (control studies), then on the last day of a 21-day course of oral estradiol-valeriate (E2) administration (2 mg daily) and finally, on the last day of a 21-day course of oral estradiol-progesterone (E2/P4) replacement (2 mg E2 and 200 mg micronized P4 daily). During all study occasions, blood was sampled at 10 min intervals for 10 h, while TRH (200 micrograms iv) was administered 8 h after initiation of blood collections. Compared to the control conditions, serum E2 and P4 concentrations markedly increased (p less than 0.001) following oral E2 or E2/P4 treatments. Total triiodothyronine (T3) and thyroxine (T4) concentrations and free T3 and T4 equivalents remained unchanged during E2 and E2/P4 regimens. In the unstimulated secretory profiles, TSH was found to be episodically released, with little interindividual variability for each study condition. Since the TSH pulse attributes (pulse amplitudes, frequencies, interpulse intervals, mean TSH concentrations, by Cluster pulse algorithm) did not significantly change during E2 and E2/P4 replacements, the episodic character of TSH secretion virtually remained unchanged by sex steroid replacements.(ABSTRACT TRUNCATED AT 250 WORDS)

Administration, Oral↗

PubMed1994

Does the gonadotropin pulsatility of postmenopausal women represent the unrestrained hypothalamic-pituitary activity?

Ovarian sex steroids profoundly modulate the gonadotropin pulsatile secretion in women. A gonadotropin pulsatility determined in the absence of any considerable ovarian sex steroid feedback, as in postmenopausal women (PMW), may thus represent the unrestrained activity of the hypothalamic-pituitary axis. We hypothesized that increases in the gonadotropin pulse frequencies and amplitudes during sex steroid replacements may be limited by those determined in the hypogonadal state of PMW. To address this assumption, we investigated the unstimulated the gonadotropin-releasing hormone (GnRH)-stimulated release of luteinizing hormone (LH) and follicle-stimulating hormone (FSH) in PMW before and during sequential ovarian sex steroid treatments. Seven PMW (mean age 59.4 years) were studied initially during unreplaced conditions (control studies), then on the last day of a 21-day course of oral estradiol valeriate (E2) administrations (2 mg daily) and, finally, on the last day of a 21-day course of oral estradiol-progesterone (E2/P4) replacements (2 mg of E2 and 200 mg of micronized P4 daily). On all study occasions, blood was drawn at 10-min intervals for 10 h and GnRH (25 micrograms iv) was administered 8 h after initiation of blood collections. Compared to control conditions, the basal serum estrogen (estrone and E2) and progesterone (P4) concentrations markedly increased (p < 0.001) following oral E2 or E2/P4 treatments. As determined by Cluster pulse algorithm, LH and FSH were found to be released episodically during each study condition. Mean LH and FSH release rates declined (P < 0.05 or less) during E2 and E2/P4 regimens.(ABSTRACT TRUNCATED AT 250 WORDS)

Androgens↗

PubMed1994

Luteinizing hormone secretory pattern before and after removal of Leydig cell tumor of the testis.

We studied the luteinizing hormone (LH) secretory pattern in three patients, aged 30, 23 and 43 years, with gynecomastia due to Leydig cell tumor of the testis, before and 6 months after unilateral orchidectomy. The results were compared to those of 11 normal fertile controls aged 20-35 years. Blood sampling was done at 20-min intervals from 22.00 h to 10.00 h. The LH data were analyzed with the Cluster analysis algorithm with "optimal parameters for LH male data" to determine the pulse interval and pulse amplitude. The Expfit program was applied to LH pulses to calculate the apparent half-life of immunoreactive LH. Before surgery, when compared to controls, the patients had a low to normal testosterone/estradiol ratio (0.053, 0.110, 0.046 vs 0.148 +/- 0.038) and mean LH levels (1.96, 3.7, 2.55 vs 4.0 +/- 1.9 IU/l), decreased pulse amplitude (2.65, 3.01, 2.21 vs 3.31 +/- 1.41 IU/l) and reduced apparent half-life of LH (74, 69, 78 vs 97 +/- 16 min). After removal of the Leydig cell tumor, the testosterone/estradiol ratio returned to the normal range (0.141, 0.177, 0.093) while an increase in mean LH levels (5.75, 7.90, 4.88 IU/l), LH pulse amplitude (3.07, 6.05, 2.86 IU/l) and apparent half-life of LH (138, 106, 104 min) was observed in all three patients. Our data indicate that endogenous hyperestrogenism in patients with Leydig cell tumor of the testis results in an inhibition of LH secretion, and suggests that such inhibition could result from a reduction in pulse amplitude and apparent half-life.

Adult↗

PubMedDate not supplied

An image retrieval system based on fractal dimension.

This paper presents a new kind of image retrieval system which obtains the feature vectors of images by estimating their fractal dimension; and at the same time establishes a tree-structure image database. After preprocessing and feature extracting, a given image is matched with the standard images in the image database using a hierarchical method of image indexing.

Abstracting and Indexing↗