Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “model selection”

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 1,135 records · Page 63Linked to original sources
PubMed2005

Predicting scorpion sting incidence in an endemic region using climatological variables.

Scorpionism is a public health problem in several regions of the world. The highest mortality, with over 1000 deaths per year, has been reported in Mexico. We analysed the significance of climatological variables to predict the incidence of scorpion stings in humans in the state of Colima (Mexico) for the years 2000-2001. The pluvial precipitation (mm), the evaporation (mm), and the mean, maximum, and minimum temperatures (degrees C) were obtained from local meteorological offices. There are approximately 3 stings/year per 1000 people in municipalities of Colima and Villa de Alvarez and about 18-30 stings/year per 1000 people in the rest of the municipalities. There is very little rain and there are few stings in the winter when the minimum temperature is below about 16 degrees C. The number of scorpion stings is independent of the actual rainfall when this is above 30 mm/month. Using multiple linear regression, we used a backward model selection procedure to estimate that the minimum temperature is correlated with scorpion sting incidence with a statistically significance of 95%. We briefly discuss the application of predictive models of scorpion sting incidence in the appropriate allocation of antivenom serum in hospital clinics.

Animals↗
PubMed2004

Dosimetry model for radioactivity localized to intestinal mucosa.

BACKGROUND: This paper provides a new model for calculating radiation-absorbed doses to the full thickness of the small and large intestinal walls, and to the mucosal layers. The model was used to estimate the intestinal radiation doses from yttrium-90-labeled-DOTA-biotin binding to NR-LU-10-streptavidin in patients. METHODS: We selected model parameters from published data and observations, and used the model to calculate energy-absorbed fractions using the EGS4 radiation transport code. We determined the cumulated (90)Y activity in the small and large intestines of patients from gamma camera images, and calculated absorbed doses to the mucosal layer and to the whole intestinal wall. RESULTS: The mean absorbed dose to the wall of the small intestine was 16.2 mGy/MBq (60 cGy/mCi) administered from (90)Y localized in the mucosa, and 70 mGy/MBq (260 cGy/mCi) to the mucosal layer within the wall. Doses to the large intestinal wall and to the mucosa of the large intestine were lower than those for the small intestine by a factor of about 2.5. These doses are greater by factors of about 5 to 6 than those that would have been calculated using the standard MIRD models that assume the intestinal activity is in the bowel contents. CONCLUSIONS: The specific uptake of radiopharmaceuticals in mucosal tissues may lead to dose-related intestinal toxicities. Tissue dosimetry at the sub-organ level is useful for a better understanding of intestinal tract radiotoxicity and associated dose-response relationships.

Humans↗
PubMed2002

Bayesian infinite mixture model based clustering of gene expression profiles.

MOTIVATION: The biologic significance of results obtained through cluster analyses of gene expression data generated in microarray experiments have been demonstrated in many studies. In this article we focus on the development of a clustering procedure based on the concept of Bayesian model-averaging and a precise statistical model of expression data. RESULTS: We developed a clustering procedure based on the Bayesian infinite mixture model and applied it to clustering gene expression profiles. Clusters of genes with similar expression patterns are identified from the posterior distribution of clusterings defined implicitly by the stochastic data-generation model. The posterior distribution of clusterings is estimated by a Gibbs sampler. We summarized the posterior distribution of clusterings by calculating posterior pairwise probabilities of co-expression and used the complete linkage principle to create clusters. This approach has several advantages over usual clustering procedures. The analysis allows for incorporation of a reasonable probabilistic model for generating data. The method does not require specifying the number of clusters and resulting optimal clustering is obtained by averaging over models with all possible numbers of clusters. Expression profiles that are not similar to any other profile are automatically detected, the method incorporates experimental replicates, and it can be extended to accommodate missing data. This approach represents a qualitative shift in the model-based cluster analysis of expression data because it allows for incorporation of uncertainties involved in the model selection in the final assessment of confidence in similarities of expression profiles. We also demonstrated the importance of incorporating the information on experimental variability into the clustering model. AVAILABILITY: The MS Windows(TM) based program implementing the Gibbs sampler and supplemental material is available at http://homepages.uc.edu/~medvedm/BioinformaticsSupplement.htm CONTACT: medvedm@email.uc.edu

Bayes Theorem↗
PubMed2006

Flexible modeling via a hybrid estimation scheme in generalized mixed models for longitudinal data.

To circumvent the computational complexity of likelihood inference in generalized mixed models that assume linear or more general additive regression models of covariate effects, Laplace's approximations to multiple integrals in the likelihood have been commonly used without addressing the issue of adequacy of the approximations for individuals with sparse observations. In this article, we propose a hybrid estimation scheme to address this issue. The likelihoods for subjects with sparse observations use Monte Carlo approximations involving importance sampling, while Laplace's approximation is used for the likelihoods of other subjects that satisfy a certain diagnostic check on the adequacy of Laplace's approximation. Because of its computational tractability, the proposed approach allows flexible modeling of covariate effects by using regression splines and model selection procedures for knot and variable selection. Its computational and statistical advantages are illustrated by simulation and by application to longitudinal data from a fecundity study of fruit flies, for which overdispersion is modeled via a double exponential family.

Animals↗
PubMed1997

Similarities and disparities between core-specific and O-side-chain-specific antilipopolysaccharide monoclonal antibodies in models of endotoxemia and bacteremia in mice.

We have previously described cross-reactive antilipopolysaccharide (anti-LPS), or anti-endotoxin, monoclonal antibodies (MAbs) which provide cross-protection in several systems of endotoxin bioactivity. The protective effects of the murine cross-reactive MAb WN1 222-5 (immunoglobulin G2a(kappa) [IgG/2a(kappa)]) and of its chimerized version, SDZ 219-800 [human IgG1(kappa)], have now been evaluated in lethality models against LPS from three different serotypes and in bacterial infection models. We confirmed the protective activity of the two MAbs in D-galactosamine-sensitized mice challenged with LPS of other E. coli serotypes (O18, O127, and O111). The protective effect correlated with the suppression of tumor necrosis factor formation. Furthermore, WN1 222-5 enhanced bacterial clearance of intravenously administered E. coli O111 bacteria, thus protecting mice from death. However, the MAbs were unable to provide protection in a peritonitis model (intraperitoneal inoculation). Our study, therefore, shows that LPS cross-reactive antibodies are capable of mediating cross-protection against LPS and bacteria but that the selected models have a clear influence on the results.

Animals↗
PubMed2006

A demonstration of modeling count data with an application to physical activity.

Counting outcomes such as days of physical activity or servings of fruits and vegetables often have distributions that are highly skewed toward the right with a preponderance of zeros, posing analytical challenges. This paper demonstrates how such outcomes may be analyzed with several modifications to Poisson regression. Five regression models 1) Poisson, 2) overdispersed Poisson, 3) negative binomial, 4) zero-inflated Poisson (ZIP), and 5) zero-inflated negative binomial (ZINB) are fitted to data assessing predictors of vigorous physical activity (VPA) among Latina women. The models are described, and analytical and graphical approaches are discussed to aid in model selection. Poisson regression provided a poor fit where 82% of the subjects reported no days of VPA. The fit improved considerably with the negative binomial and ZIP models. There was little difference in fit between the ZIP and ZINB models. Overall, the ZIP model fit best. No days of VPA were associated with poorer self-reported health and less assimilation to Anglo culture, and marginally associated with increasing BMI. The intensity portion of the model suggested that increasing days of VPA were associated with more education, and marginally associated with increasing age. These underutilized models provide useful approaches for handling counting outcomes.

Journal Article↗
PubMed2003

Ab initio construction of polypeptide fragments: Accuracy of loop decoy discrimination by an all-atom statistical potential and the AMBER force field with the Generalized Born solvation model.

The accuracy of model selection from decoy ensembles of protein loop conformations was explored by comparing the performance of the Samudrala-Moult all-atom statistical potential (RAPDF) and the AMBER molecular mechanics force field, including the Generalized Born/surface area solvation model. Large ensembles of consistent loop conformations, represented at atomic detail with idealized geometry, were generated for a large test set of protein loops of 2 to 12 residues long by a novel ab initio method called RAPPER that relies on fine-grained residue-specific phi/psi propensity tables for conformational sampling. Ranking the conformers on the basis of RAPDF scores resulted in selected conformers that had an average global, non-superimposed RMSD for all heavy mainchain atoms ranging from 1.2 A for 4-mers to 2.9 A for 8-mers to 6.2 A for 12-mers. After filtering on the basis of anchor geometry and RAPDF scores, ranking by energy minimization of the AMBER/GBSA potential energy function selected conformers that had global RMSD values of 0.5 A for 4-mers, 2.3 A for 8-mers, and 5.0 A for 12-mers. Minimized fragments had, on average, consistently lower RMSD values (by 0.1 A) than their initial conformations. The importance of the Generalized Born solvation energy term is reflected by the observation that the average RMSD accuracy for all loop lengths was worse when this term is omitted. There are, however, still many cases where the AMBER gas-phase minimization selected conformers of lower RMSD than the AMBER/GBSA minimization. The AMBER/GBSA energy function had better correlation with RMSD to native than the RAPDF. When the ensembles were supplemented with conformations extracted from experimental structures, a dramatic improvement in selection accuracy was observed at longer lengths (average RMSD of 1.3 A for 8-mers) when scoring with the AMBER/GBSA force field. This work provides the basis for a promising hybrid approach of ab initio and knowledge-based methods for loop modeling.

Algorithms↗
PubMed2003

Excitotoxic lesioning of the rat basal forebrain with S-AMPA: consequent mineralization and associated glial response.

Regional depositions of calcium within the basal ganglia, cortex, cerebellum, and white matter and at perivascular sites have been observed in several pathological conditions. These generally indicate signs of ongoing apoptosis or necrotic processes, whereby the activation of glutamate receptors causes a rise in intracellular calcium levels leading to mineralization of neurons, and ultimately to cell death. The selective degeneration of cholinergic neurons in the basal forebrain is a major neuropathological component of Alzheimer's disease, and may result in abnormal deposition of calcium. In experimental models, selective lesions of the basal forebrain can be induced by intraparenchymal infusions of excito- or immunotoxins targeting cholinergic neurons. Excitotoxic lesions are often accompanied by calcium deposition within affected areas. In a previous study we also noted the presence of unusual deposition in areas close to the site of injections following unilateral S-AMPA-induced lesions of the basal forebrain (T. Perry, H. Hodges, and J. A. Gray, 2001, Brain Res. Bull. 54, 29-48). In this paper, we have characterized these deposits histologically and evaluated the microglial (CD11b) and astrocytic (GFAP) responses at 8 and 16 weeks following lesioning of the nucleus basalis magnocellularis with S-AMPA. The resulting deposits were heterogeneous in morphology and composed primarily of calcium. Small granular deposits were detected around blood vessels, whereas larger calcospherites were situated within the parenchyma. These deposits were more widely dispersed at 16 weeks postlesioning, affected neighboring nuclei, and displayed a progressive increase in size and frequency of occurrence. However, calcification within these regions was differentially associated with microglial and astrocytic reactivity at the two time points. Both microglial and astrocytic responses were pronounced at 8 weeks, whereas at 16 weeks, astrocytic reactivity prevailed and the microglial response was markedly attenuated. Importantly, the pattern of reactivity for microglia detected at 8 weeks was specifically localized to vulnerable nucleated areas prior to their substantial accumulation of calcium deposits, which was clearly evident by 16 weeks. We suggest that the initial microglial response could be used as a selective predictor of tissue necrosis and subsequent calcification, and that astrocytes, which form a glial scar in the affected tissues, may contribute toward the buildup of calcium deposits. The functional relevance of these findings is discussed.

Animals↗
PubMedDate not supplied

[Age-period-cohort analysis of mortality caused by ischemic cardiopathy in Spain 1965-1985].

With the objective of studying the temporal evolution of ischaemic cardiopathy (IC), or coronary heart disease mortality, in Spain, we carried out a cohort analysis with conventional graphic techniques and modern statistical methods. This permits better understanding and quantification of the age-period-cohort effects and identification of the potential factors operating upon them. To this end, loglineal (Poisson regression) models were constructed of the IC mortality rates for both sexes, using the GLIM package, in which the regression coefficients are the natural Relative Risk (RR) logarithms of the various age groups (35-74 years), period of death (1970-1985) and birth cohort (1985-1960) with respect to the reference group mortality, controlled by the effect of other groups. In respect of the results, the maximum RR value corresponds to decrease year 1975, and falls progressively to 1985, though at all times remaining above the 1970 value. The effect of 1985, though less than 1980, does not present significant differences from the latter. Nevertheless, no clear cohort effect was found. As a probable explanation for the pattern observed, this would suggest recent changes in life style and in medical attention. There is a discussion of the consistency of the models selected with the graphical results and with present knowledge of the natural history of IC and with the evolution of its determining factors, together with validation of the models. In summary, the IC mortality patterns observed show an increase up to the mid-seventies, and stabilization from that date onwards, in all age and sex groups, which is consistent with an age-period effect.

Adult↗
PubMedDate not supplied

Comparing SSRI treatment costs for depression using retrospective claims data: the role of nonrandom selection and skewed data.

BACKGROUND AND OBJECTIVES: Since conventional randomized clinical trials often do not reflect the real world circumstances of prescribing behavior and patient outcomes, the use of retrospective administrative claims databases (RACD) has become more common in treatment cost comparisons among alternative pharmaceutical compounds. Several recent RACD studies have compared treatment costs for depressed patients prescribed SSRIs such as fluoxetine, sertraline and paroxetine. These cost comparisons have reached mixed conclusions. To begin to explain and reconcile the mixed SSRI cost comparison evidence, we undertake a variety of alternative multivariate analyses using a publicly available RACD. METHODS AND DATA: The 1995 to 1996 data encompasses a time period when all three SSRIs had become well-established agents. We report and compare results from multivariate linear regressions, logistic regressions, ordered probits and sample selectivity models, and examine robustness when adjustments are made for outlier observations and skewed distributions. RESULTS AND CONCLUSIONS: While choice of initial SSRI is nonrandom, the effect of sample selectivity on total depression-related and total health care expenditure is neutral across SSRIs. Although most cost measures are numerically greatest for fluoxetine, depression-related outpatient and hospitalization costs do not significantly differ by choice of initial SSRI. These findings are robust to alternative assumptions, specifications, and procedures. Antidepressant medication costs, however, are significantly higher when fluoxetine is the initial SSRI rather than sertraline or paroxetine, reflecting the larger proportion of fluoxetine patients prescribed a daily dosage of two or more capsules. Both total depression-related and total health care log-transformed costs are significantly lower for sertraline than fluoxetine.

Adult↗
PubMedDate not supplied

Meta-analysis of the Italian studies of short-term effects of air pollution (MISA), 1990-1999.

A meta-analysis of short-term effects of air pollution on health in eight Italian cities from 1990 to 1999 is presented. Death certificates and hospital admission data as well as daily concentrations of pollutants were collected. The same generalized linear model adjusted for age, day of the week, holidays, influenza epidemics, meteorological variables, and seasonality pattern was fitted to the city data. City-specific model selection was not done. In the meta-analysis, for each outcome, the city-specific estimates for each pollutant were combined using fixed and random-effects models. Hierarchical Bayesian models were use to investigate the effects of PM10 in detail. Each pollutant (SO2, NO2, CO, PM10, O3) was significantly associated with mortality for natural causes. The effect of PM10 on mortality was greater during the warm season and for elderly. A north-south gradient in risk was observed for total natural mortality. The excess risks on hospital admission were modified by deprivation score and by the NO2/PM10 ratio. Results add evidence for an association between air pollution and early mortality or morbidity and support the hypothesis of a synergism between meteorological variables and air pollution.

Adolescent↗
PubMed2006

Molecular recognition of opioid receptor ligands.

The cloning of the opioid receptors and subsequent use of recombinant DNA technology have led to many new insights into ligand binding. Instead of focusing on the structural features that lead to increased affinity and selectivity, researchers are now able to focus on why these features are important. Site-directed mutagenesis and chimeric data have often been at the forefront in answering these questions. Herein, we survey pharmacophores of several opioid ligands in an effort to understand the structural requirements for ligand binding and selectivity. Models are presented and compared to illustrate key sites of recognition for both opiate and nonopiate ligands. The results indicate that different ligand classes may recognize different sites within the receptor, suggesting that multiple epitopes may exist for ligand binding and selectivity.

Amino Acid Sequence↗
PubMedDate not supplied

Finding short DNA motifs using permuted Markov models.

Many short DNA motifs, such as transcription factor binding sites (TFBS) and splice sites, exhibit strong local as well as nonlocal dependence. We introduce permuted variable length Markov models (PVLMM) which could capture the potentially important dependencies among positions and apply them to the problem of detecting splice and TFB sites. They have been satisfactory from the viewpoint of prediction performance and also give ready biological interpretations of the sequence dependence observed. The issue of model selection is also studied.

Algorithms↗
PubMed2002

Selectivity in overlapping MAP kinase cascades.

Some protein kinases operate in more than one mitogen-activated protein-kinase (MAPK) cascade. We here address the question whether specificity of the cascades necessitates physical sequestration of these "promiscuous" kinases (e.g. by binding to scaffolds). A model is constructed, in which two MAPK cascades depend on a single MAP-kinase kinase that is not sequestered in two subpopulations. We show that in this model selective signal transduction is possible provided that there is an additional interaction at the MAP-kinase level, there is no simultaneous activation of more than one response by either signal. We discuss a number of additional interactions that can generate the selectivity, as well as some kinetic features by which this mechanism may be recognized experimentally.

Adaptation, Physiological↗
PubMed1992

Identification of surrogate rodent hosts for larval Onchocerca lienalis and induction of protective immunity in a model system.

The objectives of this project were to screen a variety of inbred rodent species and strains to determine their usefulness as surrogate hosts for the study of the early larval development of Onchocerca lienalis and then to use a selected model to study the induction of protective immunity. In the primary screen, 6 strains of mice, 5 strains of rats, jirds, and multimammate rats were tested. Animals were infected with fresh O. lienalis by subcutaneous implantation of third-stage larvae (L3) contained in diffusion chambers covered with 5.0-microns pore-size membranes. After 7 days the chambers were recovered, and larval viability and growth were assessed. Approximately one-half of inoculated larvae were recovered alive regardless of the host tested. Larvae were implanted in CBA/J and DBA/2J mice in chambers covered with membranes that prevented host cells from entering; survival and growth rates of the larvae were not altered by the absence of cells from the chambers. Cryopreserved larvae were implanted in chambers with 5.0-microns pore-size membranes in CBA/J and DBA/2J mice and Wistar Furth rats for 3-28 days. No statistically significant difference was seen in the larval recoveries on days 3-28 in all 3 hosts. Statistically significant increases in length were seen in the 3 strains from day 3 to day 14, after which growth appeared to cease. Molting from L3 to fourth-stage larvae was observed in all 3 hosts beginning on day 3, with most larvae completing the molt by day 7.(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗
PubMed2006

Automated semantic analysis of changes in image sequences of neurons in culture.

Quantitative studies of dynamic behaviors of live neurons are currently limited by the slowness, subjectivity, and tedium of manual analysis of changes in time-lapse image sequences. Challenges to automation include the complexity of the changes of interest, the presence of obfuscating and uninteresting changes due to illumination variations and other imaging artifacts, and the sheer volume of recorded data. This paper describes a highly automated approach that not only detects the interesting changes selectively, but also generates quantitative analyses at multiple levels of detail. Detailed quantitative neuronal morphometry is generated for each frame. Frame-to-frame neuronal changes are measured and labeled as growth, shrinkage, merging, or splitting, as would be done by a human expert. Finally, events unfolding over longer durations, such as apoptosis and axonal specification, are automatically inferred from the short-term changes. The proposed method is based on a Bayesian model selection criterion that leverages a set of short-term neurite change models and takes into account additional evidence provided by an illumination-insensitive change mask. An automated neuron tracing algorithm is used to identify the objects of interest in each frame. A novel curve distance measure and weighted bipartite graph matching are used to compare and associate neurites in successive frames. A separate set of multi-image change models drives the identification of longer term events. The method achieved frame-to-frame change labeling accuracies ranging from 85% to 100% when tested on 8 representative recordings performed under varied imaging and culturing conditions, and successfully detected all higher order events of interest. Two sequences were used for training the models and tuning their parameters; the learned parameter settings can be applied to hundreds of similar image sequences, provided imaging and culturing conditions are similar to the training set. The proposed approach is a substantial innovation over manual annotation and change analysis, accomplishing in minutes what it would take an expert hours to complete.

Algorithms↗
PubMed2003

Classification of dissolution profiles in terms of fractional dissolution rate and a novel measure of heterogeneity.

Dissolution profiles are classified in accordance with the shape of fractional dissolution rate function. This function is constant in time for the classical first-order model and, in this case, the dissolution is described by a monoexponential function. Therefore, any deviation of the fractional dissolution rate from the constant level suggests the presence of different (nonlinear/nonhomogenous) mechanisms in the dissolution process. The shapes of the fractional dissolution rate depend on the type of the model of dissolution; thus, classification with respect to this function is proposed as a tool for model selection. The Kullback-Leibler information distance is proposed for measuring similarity between two different drug dissolution profiles. The method is applied mainly to compare the first-order model, which characterizes a homogenous dosage form, with other common descriptors of dissolution and with experimental data.

Chemistry, Pharmaceutical↗
PubMed1999

Economics of managed care in spinal cord injury.

OBJECTIVE: To determine and describe trends in economic variables related to the care of individuals with spinal cord injury (SCI) and significant changes in these trends coincident with major developments in medical care cost control. DATA SOURCES: Data from the National Spinal Cord Injury Statistical Center (NSCISC) database were used to review the economic trends in SCI management from 1973 to 1998 and their relation to managed care and other health care cost-containment measures. A panel of SCI health care specialists was interviewed to determine the appropriate data variables to be reviewed. The Shepherd Center Care Health Management Program, Atlanta, GA, is presented as an example of a fiscally successful managed care program for patients with SCI. DATA EXTRACTION: Data from the NSCISC database for the years studied were extracted and converted to a form suitable for analysis by means of the statistical software SAS. DATA SYNTHESIS: Statistical techniques included multiple regression analysis, logistic regression analysis, and model selection methods. CONCLUSIONS: Trends in economic variables, in the care of individuals with SCI show changes coincident with the introduction of Diagnostic Related Groups (DRGs) and managed care as models for provider reimbursement. Significant changes occurred in acute care charges, rehabilitation charges, length of stay, rehospitalization 1 year postinjury, time from injury to admission to a Model System, and discharges to a nursing home.

Cost Control↗