Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Robustness”

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 685 records · Page 38Linked to original sources
PubMedDate not supplied

Robust image signal-to-noise ratio estimation using mixed Lagrange time delay estimation autoregressive model.

A new technique based on the statistical autoregressive (AR) model has recently been developed as a solution to signal-to-noise (SNR) estimation in scanning electron microscope (SEM) images. In the present study, we propose to cascade the Lagrange time delay (LTD) estimator with the AR model. We call this technique the mixed Lagrange time delay estimation autoregressive (MLTDEAR) model. In a few test cases involving different images, this model is found to present an optimum solution for SNR estimation problems under different noise environments. In addition, it requires only a small filter order and has no noticeable estimation bias. The performance of the proposed estimator is compared with three existing methods: simple method, first-order linear interpolator, and AR-based estimator over several images. The efficiency of the MLTDEAR estimator, being more robust with noise, is significantly greater than that of the other three methods.

Journal Article↗
PubMed2002

A note on ANOVA assumptions and robust analysis for a cross-over study.

Analysis of variance (ANOVA) methods are usually applied to analyse continuous data from cross-over studies. The analysis, however, may not have appropriate type I error when certain assumptions are violated. In this paper, we first clarify a conventionally minimum set of assumptions that validate the F-tests of ANOVA models for cross-over studies. We then provide a practical verification/remedy procedure based upon the theoretical developments. By applying the verification/remedy procedure, more robust analysis results can be expected from the ANOVA models.

Analysis of Variance↗
PubMed2002

Robustness of choice of number of doses for maximum likelihood estimation of the ED(50) in bioassay.

A number of studies have looked at the choice of number of doses for estimating the median effective dose (ED(50)) when the form of a dose-response curve is correctly assumed (see, for example, Müller and Schmitt, 1990). Here we address the question of the robustness of number of doses chosen using Müller and Schmitt's method when a logistic dose-response curve is incorrectly assumed. The underlying true dose-response curves considered include the probit, cubic-logistic and Aranda-Ordaz asymmetric models. The simulation results show that, for the above true dose-response curves and the uniform design density with doses spaced symmetrically or slightly asymmetrically around the assumed ED(50), choosing as many doses as possible either minimizes the mean squared error (MSE) or produces a MSE chosen to the minimum.

Computer Simulation↗
PubMed2003

Robust asymptotic sampling theory for correlations in pedigrees.

Methods to unravel the genetic determinants of non-Mendelian diseases lie at the next frontier of statistical approaches for human genetics. It is generally agreed that, before proceeding with segregation or linkage analysis, the trait under study ought to be shown to exhibit familial correlation. By coding dichotomous traits as binary variables, a single robust approach in the estimation of pedigree correlations, rather than two distinct approaches, can be used to assess the potential heritability of a trait, and, latterly, to examine the mode of inheritance. The asymptotic theory to conduct hypothesis tests and confidence intervals for correlations among different members of nuclear families is well established but is applicable only if the nuclear families are independent. As a further contribution to the literature, we derive the asymptotic sampling distribution of correlations between random variables among arbitrary pairs of members in extended families for the Pearson product-moment estimator with generalized weights. This derivation is done without assuming normality of the traits. The sampling distribution is shown to be asymptotically normal to first order, and hence large-sample hypothesis tests and confidence intervals with estimates of the variances and correlation coefficients are proposed. Discussion concludes with an example and a suggestion for future research.

ABO Blood-Group System↗
PubMed2004

Robust Bayesian decision theory applied to optimal dosage.

We give a model for constructing an utility function u(theta,d) in a dose prescription problem. theta and d denote respectively the patient state of health and the dose. The construction of u is based on the conditional probabilities of several variables. These probabilities are described by logistic models. Obviously, u is only an approximation of the true utility function and that is why we investigate the sensitivity of the final decision with respect to the utility function. We construct a class of utility functions from u and approximate the set of all Bayes actions associated to that class. Then, we measure the sensitivity as the greatest difference between the expected utilities of two Bayes actions. Finally, we apply these results to weighing up a chemotherapy treatment of lung cancer. This application emphasizes the importance of measuring robustness through the utility of decisions rather than the decisions themselves.

Antineoplastic Agents↗
PubMed2004

Robust Bayesian prediction of subject disease status and population prevalence using several similar diagnostic tests.

Sometimes several diagnostic tests are performed on the same population of subjects with the aim of assessing disease status of individuals and the prevalence of the disease in the population, but no test is a reference test. Although the diagnostic tests may have the same biological underpinnings, test results may disagree for some specific animals. In that case, it may be difficult to determine disease status for individual subjects, and consequently population prevalence estimation becomes difficult. In this paper, we propose a robust method of estimating disease status and prevalence that uses heavy-tailed sampling distributions in a hierarchical model to protect against the influence of conflicting observations on inferences. If a subject has a test outcome that is discordant with the other test results then it is downweighted in diagnosing a subject's disease status, and for estimating disease prevalence. The amount of downweighting depends on the degree of conflict among the test results for the subject.

Animals↗
PubMed2006

A robust approach to t linear mixed models applied to multiple sclerosis data.

We discuss a robust extension of linear mixed models based on the multivariate t distribution. Since longitudinal data are successively collected over time and typically tend to be auto-correlated, we employ a parsimonious first-order autoregressive dependence structure for the within-subject errors. A score test statistic for testing the existence of autocorrelation among the within-subject errors is derived. Moreover, we develop an explicit scoring procedure for the maximum likelihood estimation with standard errors as a by-product. The technique for predicting future responses of a subject given past measurements is also investigated. Results are illustrated with real data from a multiple sclerosis clinical trial.

Bayes Theorem↗
PubMed2006

Robust estimation in mixed linear models with non-monotone missingness.

We introduce a model to account for abrupt changes among repeated measures with non-monotone missingness. Development of likelihood inferences for such models is hard because it involves intractable integration to obtain the marginal likelihood. We use hierarchical likelihood to overcome such difficulty. Abrupt changes among repeated measures can be well described by introducing random effects in the dispersion. A simulation study shows that the resulting estimator is efficient, robust against misspecification of fatness of tails. For illustration we use a schizophrenic behaviour data presented by Rubin and Wu.

Computer Simulation↗
PubMed2007

Estimation of the proportion of overweight individuals in small areas - a robust extension of the Fay-Herriot model.

Hierarchical model such as Fay-Herriot (FH) model is often used in small area estimation. The method might perform well overall but is vulnerable to outliers. We propose a robust extension of the FH model by assuming the area random effects follow a t distribution with an unknown degrees-of-freedom parameter. The inferences are constructed using a Bayesian framework. Monte Carlo Markov Chain (MCMC) such as Gibbs sampling and Metropolis-Hastings acceptance and rejection algorithms are used to obtain the joint posterior distribution of model parameters. The procedure is used to estimate the county-level proportion of overweight individuals from the 2003 public-use Behavioral Risk Factor Surveillance System (BRFSS) data. We also discuss two approaches for identifying outliers in the context of this application.

Adolescent↗
PubMed2001

Power comparison of robust approximate and non-parametric tests for the analysis of cross-over trials.

The main advantage of cross-over designs in practice is the use of a smaller number of subjects to produce treatment comparisons with sufficient precision. Bellavance and Tardif proposed a non-parametric approach to test the hypotheses of direct treatment and carry-over effects for the three-treatment three-period and six sequences cross-over design and showed the high asymptotic efficiency of their approach relative to the classical F-test based on ordinary least squares (OLS). In a more recent paper, Ohrvik suggested another non-parametric method for the analysis of cross-over trials. The power of these two non-parametric approaches is evaluated for small sample sizes via simulations, and compared to the power of the usual analysis of variance model based on OLS and a modified F-test approximation that take into account the correlation structure of the repeated measurements within subjects. Different covariance structures, sample sizes, and probability distributions for the responses, namely normal and gamma, are used in the simulations to evaluate the power and robustness of these different methods of analysis.

Analysis of Variance↗
PubMed1993

Robustness of G protein changes in cocaine sensitization shown with immunoblotting.

Daily cocaine administration has been shown to alter G proteins in mesolimbic nuclei, and these changes have been implicated in the initiation and expression of behavioral sensitization. To evaluate the robustness of changes in G proteins induced by daily cocaine treatments capable of producing behavioral sensitization, the levels of Gi1 alpha, Gi2 alpha, Go alpha, Gs alpha, and G beta protein were measured by immunoblotting at 1 hr after an acute injection of cocaine or saline given 1 or 14 days following the last injection of daily cocaine or saline. A significant decline in Gi1 alpha was seen in the nucleus accumbens at 14 days following daily cocaine administration regardless of whether they received an acute challenge with cocaine or saline 1 hr prior to decapitation. No alterations were observed in the ventral tegmental area, substantia nigra, dorsolateral striatum, or prefrontal cortex in the levels of Gi1 alpha, Gi2 alpha, or Go alpha. No change in G protein immunoreactivity was measured in the nucleus accumbens or ventral tegmental area of rats decapitated 1 hr after discontinuing daily cocaine. The possibility that a long-term change in Gi1 alpha in the nucleus accumbens may be related to cocaine-induced behavioral sensitization is discussed.

Animals↗
PubMed2006

DNA replication in the fission yeast: robustness in the face of uncertainty.

DNA replication, the process of duplication of a cell's genetic content, must be carried out with great precision every time the cell divides, so that genetic information is preserved. Control mechanisms must ensure that every base of the genome is replicated within the allocated time (S-phase) and only once per cell cycle, thereby safeguarding genomic integrity. In eukaryotes, replication starts from many points along the chromosome, termed origins of replication, and then proceeds continuously bidirectionally until an opposing moving fork is encountered. In contrast to bacteria, where a specific site on the genome serves as an origin in every cell division, in most eukaryotes origin selection appears highly stochastic: many potential origins exist, of which only a subset is selected to fire in any given cell, giving rise to an apparently random distribution of initiation events across the genome. Origin states change throughout the cell cycle, through the ordered formation and modification of origin-associated multisubunit protein complexes. State transitions are governed by fluctuations of cyclin-dependent kinase (CDK) activity and guards in these transitions ensure system memory. We present here DNA replication dynamics, emphasizing recent data from the fission yeast Schizosaccharomyces pombe, and discuss how robustness may be ensured in spite of (or even assisted by) system randomness.

DNA Replication↗
PubMed2002

Enveloped virus inactivation by caprylate: a robust alternative to solvent-detergent treatment in plasma derived intermediates.

Solvent-detergent treatment, although used routinely in plasma product processing to inactivate enveloped viruses, substantially reduces product yield from the human plasma resource. To improve yields in plasma product manufacturing, a new viral reduction process has been developed using the fatty acid caprylate. As licensure of plasma products warrants thorough evaluation of pathogen reduction capabilities, the present study examined susceptibility of enveloped viruses to inactivation by caprylate in protein solutions with varied pH and temperature. In the immunoglobin-rich solutions from Cohn Fraction II+III, human immunodeficiency virus, Type-1, bovine viral diarrhea virus (BVDV), and pseudorabies virus were inactivated by caprylate concentrations of >/=9 mM, >/=12 mM, and >/=9 mM, respectively. Compared to solvent-detergent treatment, BVDV inactivation in Fraction II+III solution was significantly faster (20-60 fold) using 16 mM caprylate. Caprylate-mediated inactivation of BVDV was not noticeably affected by temperature within the range chosen manufacturing the immunoglobulin product. In Fraction II+III solutions, IgG solubility was unaffected by </=19 mM caprylate. In albumin solution from Cohn supernatant IV-1, 40 mM caprylate rapidly inactivated BVDV, demonstrating versatility in inactivating enveloped viruses potentially present in other protein solutions. Our data show that caprylate is a robust enveloped virus inactivating agent for immunoglobulins and albumin which may potentially be utilized for other proteins; viral inactivation was not adversely affected by protein content and the buffer composition conditions evaluated. Within the parameters examined, caprylate inactivation of enveloped viruses provided comparable activity or advantages relative to the current, standard solvent-detergent treatment.

Albumins↗
PubMed2002

A case study of logistic QSAR modeling methods and robustness tests.

Acute toxicity (15-min EC(50)) determination of 16 substituted naphthalene compounds to Photobacterium phosphoreum was undertaken according to the standard procedures, while the effects of molecular structures of selected compounds on their toxicity to test microorganisms were logistically conducted using the quantitative structure-activity relationship (QSAR) technique. The relationship was developed as -log EC(50)=5.5916 (+/-0.1189)-7.4893(+/-0.4900)qH(+)+0.7771(+/-0.0619E(lumo)+0.0088(+/-0.0009)alpha(N=16, R(2)(adj)=0.9698, SE=0.0892, P=0.0000). The cluster analyses of individual structure descriptors as well as the quality control chart and Monte Carlo simulation test indicated that the prediction model was reasonable and robust even if tested with several different methods. Furthermore, the quantum chemical parameters entering into the QSAR model were used to discuss the possible toxicity pathways, and the results revealed that the selected compounds were reactive and their toxicity behaviors were complex processes containing physical partition stages as well as biochemical reaction stages.

Lethal Dose 50↗
PubMed2000

Robust regeneration of CNS axons through a track depleted of CNS glia.

Transected CNS axons do not regenerate spontaneously but may do so if given an appropriate environment through which to grow. Since molecules associated with CNS macroglia are thought to be inhibitory to axon regeneration, we have tested the hypothesis that removing these cell types from an area of brain will leave an environment more permissive for axon regeneration. Adult rats received unilateral knife cuts of the nigrostriatal tract and ethidium bromide (EB) was used to create a lesion devoid of astrocytes, oligodendrocytes, intact myelin sheaths, and NG2 immunoreactive cells from the site of the knife cut to the ipsilateral striatum (a distance of 6 mm). The regenerative response and the EB lesion environment was examined with immunostaining and electron microscopy at different timepoints following surgery. We report that large numbers of dopaminergic nigral axons regenerated for over 4 mm through EB lesions. At 4 days postlesion dopaminergic sprouting was maximal and the axon growth front had reached the striatum, but there was no additional growth into the striatum after 7 days. Regenerating axons did not leave the EB lesion to form terminals in the striatum, there was no recovery of function, and the end of axon growth correlated with increasing glial immunoreactivity around the EB lesion. We conclude that the removal of CNS glia promotes robust axon regeneration but that this becomes limited by the reappearance of nonpermissive CNS glia. These results suggest, first, that control of the glial reaction is likely to be an important feature in brain repair and, second, that reports of axon regeneration must be interpreted with caution since extensive regeneration can occur simply as a result of a major glia-depleting lesion, rather than as the result of some other specific intervention.

Animals↗
PubMed1997

Contact area difference (CAD): a robust measure to evaluate accuracy of protein models.

A simple unified measure to evaluate the accuracy of three-dimensional atomic protein models is proposed. This measure is a normalized sum of absolute differences of residue-residue contact surface areas calculated for a reference structure and a model. It employs more rigorous quantitative evaluation of a contact than previously used contact measures. We argue that the contact area difference (CAD) number is a robust single measure to evaluate protein structure predictions in a wide range of model accuracies, from ab initio and threading models to models by homology, since it reflects both backbone topology and side-chain packing, is smooth, continuous and threshold-free, is not sensitive to typical crystallographic errors and ambiguities, adequately penalizes domain and/or secondary structure rearrangements and protein plasticity, and has consistent linear and matrix representations for more detailed analysis. The CAD quality of crystallographic structures, NMR structures, models by homology, and unfolded and misfolded structures is evaluated. It is shown that the CAD number discriminates between models better than Cartesian root-mean-square deviation (cRMSD). Structural variability of the NMR structures was found to be three times larger than deformations of crystallographic structures in different packing environments.

Amino Acids↗
PubMed2001

Target recognition by EcoKI: the recognition domain is robust and restriction-deficiency commonly results from the proteolytic control of enzyme activity.

We report a genetic and biochemical analysis of a target recognition domain (TRD) of EcoKI, a type I restriction and modification enzyme. The TRDs of type I R-M systems are within the specificity subunit (HsdS) and HsdS confers sequence specificity to a complex endowed with both restriction and modification activities. Random mutagenesis has revealed that most substitutions within the amino TRD of EcoKI, a region comprising 157 amino acid residues, have no detectable effect on the phenotype of the bacterium, even when the substitutions are non- conservative. The structure of the TRD appears to be robust. All but one of the six substitutions that confer a restriction-deficient, modification-deficient (r(-)m(-)) phenotype were found to be in the interval between residues 80 and 110, a region predicted by sequence comparisons to form part of the protein-DNA interface. Additional site-directed mutations affecting this interval commonly impair both restriction and modification. However, we show that an r(-) phenotype cannot be taken as evidence that the EcoKI complex lacks endonuclease activity; in response to even a slightly impaired modification efficiency, the endonuclease activity of EcoKI is destroyed by a process dependent upon the ClpXP protease. Enzymes from mutants with an r(-)m(-) phenotype commonly retain some sequence-specific activity; methylase activity can be detected on hemimethylated DNA substrates and residual endonuclease activity is implied whenever the viability of the r(-)m(-) bacterium is dependent on ClpXP. Conversely, the viability of ClpX(-) r(-)m(-) bacteria can be used as evidence for little, or no, endonuclease activity. Of 14 mutants with an r(-)m(-) phenotype, only six are viable in the absence of ClpXP. The significance of four of the six residues (G91, G105, F107 and G141) is enhanced by the finding that even conservative substitutions for these residues impair modification, thereby conferring an r(-)m(-) phenotype.

Adenosine Triphosphatases↗
PubMed2002

A robust technique for two-dimensional separation of undistorted chemical-shift anisotropy powder patterns in magic-angle-spinning NMR.

A robust magic-angle-spinning experiment for separating undistorted, quasi-static chemical-shift powder patterns is presented. It is derived from the technique of R. Tycko, G. Dabbagh, and P. Mirau (1989, J. Magn. Reson. 85, 265), but uses 360 degrees instead of 180 degrees pulses. In combination with a suitable phase sequence, this makes the spectral patterns very insensitive to pulse-length errors and other imperfections, as demonstrated both experimentally and theoretically. This method, termed separation of undistorted powder patterns by effortless recoupling (SUPER), can be used at standard spinning speeds, between 2.5 and 5 kHz, and with standard radiofrequency power levels (in particular, for protons, a decoupling field strength gammaB(1)/2pi of less than 80 kHz). No significant artifacts are observed even for samples extending to the ends of the radiofrequency coil. The method has been applied to samples containing various sp(2)- and sp(3)-hybridized carbon sites. Even for the methylene groups in polyethylene, a system with very strong C-H and H-H dipolar couplings and only moderate chemical-shift anisotropy (CSA), a useful CSA powder pattern has been obtained. With a CSA scaling factor of 0.155, accuracies of +/-5, +/-3, and +/-1 ppm of the principal values can be achieved for protonated aromatic carbons, unprotonated sp(2)-hybridized groups, and aliphatic sites, respectively. Examples of CSA-based assignment of COOC vs other COO or CON groups, and of aromatic vs olefinic C=C carbons are shown, for both small molecules and polymers.

Journal Article↗