Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “network analysis”

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 1,423 records · Page 79Linked to original sources
PubMed2004

Simulation of 13C nuclear magnetic resonance spectra of lignin compounds using principal component analysis and artificial neural networks.

Theoretical models relating atom-based structural descriptors to 13C NMR chemical shifts were used to accurately simulate 13C NMR spectra of lignin model compounds (poly-substituted phenols). The structure-activity relationship (SAR) studies for 15 lignins using pattern recognition methods of principal component analysis (PCA) and artificial neural networks (ANNs) were performed in this work. The most important parameters affecting the 13C chemical shifts of different carbons were descriptors consisting of the charge density of the atoms at different distances from the center carbon. Among the large number of parameters, these descriptors were selected using PCA and were used as ANN input. The least square regression analyses of the results indicate correlation coefficient (R) values in excess of 0.983 for the total data set.

Carbon Isotopes↗
PubMed2006

A fractal analysis approach to viscoelasticity of physically cross-linked barley beta-glucan gel networks.

The structure and gelation kinetics of mixed linkage barley beta-glucans of varying Mw have been investigated. The fractal concept has been applied to describe the structure development of barley beta-glucan gels using a scaling model and dynamic rheometry data. The model supports that the gel structure consists of fractal clusters that upon aggregation lead to a three-dimensional network. The analysis showed that with increasing Mw a denser (more packed) network is formed as indicated by the corresponding fractal dimension (df) values. The microelastic parameter of the model, alpha, showed that all gel structures were in the transition regime implying structural reordering upon ageing. The description of the microstructure as a fractal network seems to be able to explain syneresis and other observations from large deformation testing of such systems. The molecular treatment of the gelation kinetics suggests that the gelling behavior is governed by the probability of collision of chain fragments with consecutive cellotriosyl units. This is greater for small chains due to their higher diffusion rates, for chains having lower amounts of cellulose like fragments and finally for those showing smaller degree of intrachain interactions. As a result, the faster gelling systems exhibit lower fractal dimensionality (more disordered systems) something that is in accordance with current kinetic theories.

Algorithms↗
PubMed1989

The spectrin network as a barrier to lateral diffusion in erythrocytes. A percolation analysis.

The spectrin network on the cytoplasmic surface of an erythrocyte can be modeled as a triangular lattice of spectrin tetramers (Tsuji, A., and S. Ohnishi, 1986. Biochemistry. 25:6133-6139). The tetramers act as barriers to protein diffusion, while dissociated dimer pairs, single dimers, and missing tetramers do not. Diffusion in the presence of these barriers is shown to be equivalent to bond percolation on the honeycomb lattice. Monte Carlo calculations for this system then yield the relative diffusion constant of a mobile integral protein as a function of the fraction of spectrin tetramers. At high concentrations of spectrin tetramer, long-range diffusion is blocked, but short-range diffusion is still possible. Monte Carlo calculations yield the average distance over which short-range diffusion can occur, as a function of the fraction of spectrin tetramers. Applications to erythrocyte development and hereditary hemolytic anemia are discussed.

Algorithms↗
PubMed2000

Computerized classification of corpus cavernosum electromyogram signals by the use of discriminant analysis and artificial neural networks to support diagnosis of erectile dysfunction.

Corpus cavernosum electromyogram (CC-EMG) provides diagnostic information on cavernous autonomic innervation and a measure of the degree to which the cavernous smooth muscle cells are intact. The complicated CC-EMG is evaluated and used in the diagnosis of patients suffering from erectile dysfunction. The evaluation procedure has been simplified by applying digital signal processing techniques. Since mathematically-based interpretations require quantitative data, spectral analysis was performed. The derived biosignals were analyzed by fast Fourier transform (FFT). Besides various other spectral parameters, specific frequency bands were determined in the power spectrum using factor analysis. The parameters were used for the computerized classification of normal and pathological CC-EMG data and the classification was performed using two independent methods: discriminant analysis (DA) and artificial neural networks (ANN). A medical expert analyzed a total of 200 CC-EMG recordings from patients with and without erectile dysfunction and separated these into normal (136) and pathological (64) cases. Although each independent method had already resulted in a relatively high number of correct classifications, the classification success rate could be slightly improved by using a combination of both classification methods. A total of 72.79% and 77.94% were successfully classified using DA and ANN, respectively. The combination of both methods increased the classification success to 80.15%. The results of this study enabled impartial evaluation of the CC-EMG signals for clinical diagnostic purposes of erectile dysfunction. This method provided an objective and easy way to analyze the CC-EMG. Furthermore, this results in patient diagnosis becoming an easier task for less experienced doctors, since little knowledge of the raw signal is needed.

Algorithms↗
PubMed1983

Three-dimensional structure of actin filaments and of an actin gel made with actin-binding protein.

Purified muscle actin and mixtures of actin and actin-binding protein were examined in the transmission electron microscope after fixation, critical point drying, and rotary shadowing. The three-dimensional structure of the protein assemblies was analyzed by a computer-assisted graphic analysis applicable to generalized filament networks. This analysis yielded information concerning the frequency of filament intersections, the filament length between these intersections, the angle at which filaments branch at these intersections, and the concentration of filaments within a defined volume. Purified actin at a concentration of 1 mg/ml assembled into a uniform mass of long filaments which overlap at random angles between 0 degrees and 90 degrees. Actin in the presence of macrophage actin-binding protein assembled into short, straight filaments, organized in a perpendicular branching network. The distance between branch points was inversely related to the molar ratio of actin-binding protein to actin. This distance was what would be predicted if actin filaments grew at right angles off of nucleation sites on the two ends of actin-binding protein dimers, and then annealed. The results suggest that actin in combination with actin-binding protein self-assembles to form a three-dimensional network resembling the peripheral cytoskeleton of motile cells.

Actins↗
PubMedDate not supplied

The applicability of recurrent neural networks for biological sequence analysis.

Selection of machine learning techniques requires a certain sensitivity to the requirements of the problem. In particular, the problem can be made more tractable by deliberately using algorithms that are biased toward solutions of the requisite kind. In this paper, we argue that recurrent neural networks have a natural bias toward a problem domain of which biological sequence analysis tasks are a subset. We use experiments with synthetic data to illustrate this bias. We then demonstrate that this bias can be exploitable using a data set of protein sequences containing several classes of subcellular localization targeting peptides. The results show that, compared with feed forward, recurrent neural networks will generally perform better on sequence analysis tasks. Furthermore, as the patterns within the sequence become more ambiguous, the choice of specific recurrent architecture becomes more critical.

Algorithms↗
PubMedDate not supplied

PACSPulse: a web-based DICOM network traffic monitor and analysis tool.

PACSPulse, an open-source tool, was developed to identify and analyze the performance bottlenecks of picture archiving and communication systems (PACS). PACSPulse provides a graphical Web interface for straightforward analysis of PACS performance on the basis of data acquired by tracking usage by network, server, workstation, type of traffic, and time of day. The PACS archive logs performance and usage data on image traffic being sent to it from the imaging units and study data requested by users. The performance log is sent via file transfer protocol (FTP) to a separate server for analysis. The data are parsed and sent to a database server connected to a Web server. The Web site is used to depict trends in the performance of the entire system to detect signs of degradation. The system was built entirely of open-source components for the operating system, database, charting tool, and Web server. Performance monitoring is an essential tool for analyzing, understanding, and predicting the performance characteristics of a PACS.

Humans↗
PubMed2002

Metallic content of wines from the Canary Islands (Spain). Application of artificial neural networks to the data analysis.

Eleven elements, K, Na, Ca, Mg, Fe, Cu, Zn, Mn, Sr, Li and Rb, were determined in dry and sweet wines bearing the denominations of origin of El Hierro, La Palma and Lanzarote islands (Canary Islands, Spain). Analyses were performed by flame atomic absorption spectrophotometry, with the exceptions of Li and Rb for which flame atomic emission spectrophotometry was used. The content in copper and iron did not present risks of cases. All samples presented a copper and zinc content below the maximum amount recommended by the Office International de la Vigne et du Vin (OIV) for these elements. Significant differences in the metallic content were found among the different islands. Thus, Lanzarote presented the highest mean content in sodium and lithium and the lowest mean content in rubidium, and La Palma presented the highest mean content in strontium and rubidium. Sweet wines from La Palma, elaborated as naturally sweet with over-ripe grapes, presented mean contents significantly higher with regard to dry wines from the same island in the majority of the analysed elements. Cluster analysis and Kohonen self-organising maps showed differences in wines according to the island of origin and the ripening state of the grapes. Back-propagation artificial neural networks showed better prediction ability than stepwise linear discriminant analysis.

Atlantic Islands↗
PubMed1996

Automated classification of renal interstitium and tubules by local texture analysis and a neural network.

OBJECTIVE: To segment renal interstitial space in order to automatically quantify renal cortical interstitial volume fraction (Vvint/cortex) by means of image analysis techniques. STUDY DESIGN: The study group consisted of 35 renal biopsies with different degrees of chronic interstitial damage. Biopsies were stained with Sirius red and digitized under polarized light. Two methods were employed to segment interstitial space: (1) interstitial bright particles were thresholded, and afterwards interstitial space was reconstructed with a morphologic operation, and (2) the texture of the surroundings of each pixel was quantified by means of local granulometry, and this information was employed as the input of a neural network in order to classify interstitial and tubular pixels. RESULTS: The correlation between Vvint/cortex obtained manually and both methods was r = .92. The first method produced some deformation of tubular contours and underestimated Vvint/cortex (beta = .70) when compared to the second approach (beta = .95) (P < .05). CONCLUSION: Two different algorithms based on image analysis techniques allow the classification of renal interstitial and tubular structures and consequently allow the automated and precise estimation of renal Vvint/cortex.

Adult↗
PubMed1996

An electrical network model of intracranial arteriovenous malformations: analysis of variations in hemodynamic and biophysical parameters.

The propensity of intracranial arteriovenous malformations (AVMs) to hemorrhage is correlated significantly with their hemodynamic features. Biomathematical models offer a theoretical approach to analyse complex AVM hemodynamics, which otherwise are difficult to quantify, particularly within or in close proximity to the nidus. Our purpose was to investigate a newly developed biomathematical AVM model based on electrical network analysis in which morphological, biophysical, and hemodynamic characteristics of intracranial AVMs were replicated accurately. Several factors implemented into the model were altered systematically to study the effects of a possible wide range of normal variations in AVM hemodynamic and biophysical parameters on the behavior of this model and its fidelity to physiological reality. The model represented a complex, noncompartmentalized AVM with four arterial feeders, two draining veins, and a nidus consisting of 28 interconnected plexiform and fistulous components. Various clinically-determined experimentally-observed, or hypothetically-assumed values for the nidus vessel radii (plexiform: 0.01 cm-0.1 cm; fistulous: 0.1 cm-0.2 cm), mean systemic arterial pressure (71 mm Hg-125 mm Hg), mean arterial feeder pressures (21 mm Hg-80 mm Hg), mean draining vein pressures (5 mm Hg-23 mm Hg), wall thickness of nidus vessels (20 microns-70 microns), and elastic modulus of nidus vessels (1 x 10(4) dyn/cm2 to 1 x 10(5) dyn/cm2) were used as normal or realistic ranges of parameters implemented in the model. Using an electrical analogy of Ohm's law, flow was determined based on Poiseuille's law given the aforementioned pressures and resistance of each nidus vessel. Circuit analysis of the AVM vasculature based on the conservation of flow and voltage revealed the flow rate through each vessel in the AVM network. An expression for the risk of AVM nidus rupture was derived based on the functional distribution of the critical radii of component vessels. The two characteristics which were used to judge the fidelity of the theoretical performance of the AVM model against the physiological one of human AVMs were total volumetric flow through the AVM (< or = 900 ml/min), and its risk of rupture (< 100%). Applying these criteria, a series of 216 (out of 260) AVM models using different combinations of these hemodynamic and biophysical parameters resulted in a physiologically-realistic conduct of the model (yielding a total flow through the AVM model varying from 449.9 ml/min to 888.6 ml/min, and a maximum risk of rupture varying from 26.4 to 99.9%). The described novel biomathematical model characterizes the transnidal and intranidal hemodynamics of an intracranial AVM more accurately than previously possible. A wide range of hemodynamic and biophysical parameters can be implemented in this AVM model to result in simulation of human AVMs with differing characteristics (e.g. low-flow and high-flow AVMs). This experimental model should serve as a useful research tool for further theoretical investigations of a variety of intracranial AVMs and their hemodynamic sequelae.

Analysis of Variance↗
PubMed1998

The role of social determinants on men's and women's mobility in Italy. A comparison of discriminant analysis and artificial neural networks.

The paper focuses on the role of the spouse's occupation as a resource for mobile individuals, from the perspective that social positions are held by families, rather than by individuals. Three groups are confronted in terms of the role of the key variables and other relevant factors: men whose spouse does not have a paid job (group 1), men and women whose spouse has a paid job (group 2 and 3). The data set is provided by the national survey on social mobility in Italy, carried out in 1985; social achievements of members of the three groups are considered, including social origins and destinations, social position corresponding to respondent's first job, cultural background (educational achievement of respondent's father and mother found), respondent's education and spouse's social position. The techniques used are discriminant analysis and back propagation Neural Networks. Both techniques traced a clear boundary between group 1 and groups 2 and 3, which were discriminated mainly on the basis of the spouse's occupation; Artificial Neural Networks reached better classification results and allowed a deeper insight into the nonlinear effects of the discriminating variables for the three groups.

Artificial Intelligence↗
PubMed1996

Descriptors and techniques for quantitative structure-biodegradability studies.

Biodegradation occurs mainly through microbial enzyme attack, and enzyme-catalysed reactions are known to depend on hydrophobic, electronic and steric effects. However, most QSBR studies involve correlation with a single parameter, but there is no consistency regarding the class of parameter. This suggests that biodegradation occurs through a range of different mechanisms. A few QSBR studies have reported the need to include more than one class of parameter in correlations. As well as linear regression analysis, other correlation methods have been used in QSBR investigations. These include discriminant analysis, neural networks and comparative molecular field analysis (CoMFA).

Bacteria↗
PubMed2006

Graph theoretic properties of networks formed by the Delaunay tessellation of protein structures.

The Delaunay tessellation of several sets of real and simplified model protein structures has been used to explore graph theoretic properties of residue contact networks. The system of contacts defined by residues joined by edges in the Delaunay simplices can be thought of as a graph or network and analyzed using techniques from elementary graph theory and the theory of complex networks. Such analysis indicates that protein contact networks have small world character, but technically are not small world networks. This approach also indicates that networks formed by native structures and by most misfolded decoys can be differentiated by their respective graph properties. The characteristic features of residue contact networks can be used for the detection of structural elements in proteins, such as the ubiquitous closed loops consisting of 22-32 consecutive residues, where terminal residues are Delaunay neighbors.

Binding Sites↗
PubMed2002

Epicenter location by analysis of interictal spikes: a case study for the use of artificial neural networks in biomedical engineering.

Artificial neural network (ANN) technology is finding increasing application in medicine and biomedical engineering. This paper supplies necessary background in ANN technology for researchers unfamiliar with this rapidly emerging discipline. This introduction to ANN application is cast in the context of epileptic seizure epicenter location. This is a very real problem faced by neurosurgeons every day. Precise location of the area of excision is currently determined with a network of surgically implanted subdural electrodes. This means that the cure entails two surgical procedures: one to implant the electrode array that precisely locates the epicenter, and another to remove the epicenter. This paper outlines an experimental diagnostic software system (DSS) that uses artificial neural network (ANN) analysis of magnetoencephalographic (MEG) data to eliminate the first of these surgical procedures. The MEG recording is a quick and painless process that requires no surgery. This approach has the potential to save time, reduce patient discomfort, and eliminate a painful and potentially dangerous surgical step in the treatment procedure.

Biomedical Engineering↗
PubMed2005

Network structure and attitudes toward collaboration in a community partnership for diabetes control on the US-Mexican border.

PURPOSE: This study seeks to provide an examination of a health policy network operating in a single, small community along the US-Mexican border. The purpose of the paper is to discuss why and how this network evolved, and then to present findings on how the network was structured. Analysis will focus especially on agency involvement, or "embeddedness" in the network, and its relationship to attitudes held by network members regarding trust, reputation, and perceived benefit. DESIGN/METHODOLOGY/APPROACH: Data were collected from 15 public and nonprofit agencies trying to work collaboratively to influence local policy and services regarding the prevention of obesity-related chronic disease, especially diabetes. Embeddedness was measured in three different ways and both confirmed and unconfirmed networks were assessed. Network analysis methods were utilized as well as nonparametric correlation statistics. FINDINGS: The network was found to be densely connected through unconfirmed linkages, but much less so when these links were confirmed. Strongest findings were found for shared information. Measures of agency embeddedness in the network were strong predictors of agency reputation, but findings for trust and perceived benefit were generally weak. ORIGINALITY/VALUE: From a practice perspective, the study points to the problems in building and sustaining community-based chronic disease health networks, especially in a small community with substantial health needs. The research also contributes to theory on embeddedness and to methodology for collecting and analyzing data on community health networks.

Adult↗
PubMed1998

Analysis of atypical squamous (glandular) cells of undetermined significance smears by neural network-directed review.

OBJECTIVE: The objective of this study was to evaluate the value of neural network-directed review of smears determined to contain atypical squamous (glandular) cells of undetermined significance to identify those cases most likely to be associated with cervical intraepithelial neoplasia. METHODS: One hundred sixty smears reported as atypical squamous (glandular) cells of undetermined significance on patients having colposcopy and directed biopsy within 1 year of the smear were identified. The smears were subjected to a neural network-directed review and classified according to findings on this review. The latter findings were related to those obtained on cervical biopsy. RESULTS: One hundred sixty smears originally reported as atypical squamous (glandular) cells of undetermined significance were subjected to neural network-directed review. The smears were upgraded in 20.6% of cases. Ninety-one patients were found to have normal biopsies, and 69 had biopsies reported as abnormal. Of the smears in patients with abnormal biopsies, 37.7% were upgraded, whereas only 7.7% of smears from those with normal biopsies were upgraded (P < .001). Nine patients were found to have cervical intraepithelial neoplasia-3 on biopsy. Six of the nine smears (66.7%) taken on these patients were upgraded. CONCLUSION: Neural network-directed analysis of smears conventionally diagnosed as atypical squamous (glandular) cells of undetermined significance will reveal findings suggesting a squamous intraepithelial lesion in a significant number of cases. This approach requires further study because it is a relatively cost-effective means of triaging patients with a cytologic diagnosis of atypical squamous (glandular) cells of undetermined significance.

Adolescent↗
PubMedDate not supplied

Genetic network identification by high density, multiplexed reversed transcriptional (HD-MRT) analysis in steroidogenic axis model cell lines.

Transcriptional network analysis in steroidogenic axis cell lines requires an understanding of cellular network composition and complexity. Previous studies have shown that absence of transcriptional network components in a cell line compromises that cell line's functional capacity for transcriptional regulation. Our goal was to analyze qualitatively steroidogenic axis-derived cell lines' expression of a putative transcriptional network involved in human and mouse development. To pursue this analysis we used Northern blots and a high density-multiplexed reverse transcription-polymerase chain reaction (HD-MRT-PCR) approach. Our results revealed that, while some members of this putative network were universally expressed, only a minority of the non-constitutive targeted transcripts were present in any single line. Based on our data and previously published results for contextual expression of these transcription factors, a model was constructed possessing the topology suggestive of a scale-free network: certain network members were highly connected nodes and would represent critical sites of vulnerability. The importance of these highly connected nodes for network function is supported by the severe phenotypes exhibited by human patients and animal models when these genes are mutated. We conclude that knowledge of network composition in specific cell lines is essential for their use as models to investigate functional interactions within selected subnetworks.

Alternative Splicing↗
PubMed2000

The non-grid technique for modeling 3D QSAR using self-organizing neural network (SOM) and PLS analysis: application to steroids and colchicinoids.

A novel method for modeling 3D QSAR has been developed. The method involves a multiple training of a series of self-organizing networks (SOM). The obtained networks have been used for processing the data of one reference molecule. A scheme for the analysis of such data with the PLS analysis has been proposed and tested using the steroids data with corticosteroid binding globulin (CBG) affinity. The predictivity of the CBG models measured with the SDEP parameter is among the best one reported. Although 3-D QSAR models for colchicinoid series is far less predictive, it allows for a discussion on the relative influence of the structural motifs of these compounds.

Binding Sites↗