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 1,063 records · Page 59Linked to original sources
PubMed1996

Mixed Poisson regression models with covariate dependent rates.

This paper studies a class of Poisson mixture models that includes covariates in rates. This model contains Poisson regression and independent Poisson mixtures as special cases. Estimation methods based on the EM and quasi-Newton algorithms, properties of these estimates, a model selection procedure, residual analysis, and goodness-of-fit test are discussed. A Monte Carlo study investigates implementation and model choice issues. This methodology is used to analyze seizure frequency and Ames salmonella assay data.

Algorithms↗
PubMed2004

[Statistical models for spatial analysis in parasitology].

The simplest way to study the spatial pattern of a disease is the geographical representation of its cases (or some indicators of them) over a map. Maps based on raw data are generally "wrong" since they do not take into consideration for sampling errors. Indeed, the observed differences between areas (or points in the map) are not directly interpretable, as they derive from the composition of true, structural differences and of the noise deriving from the sampling process. This problem is well known in human epidemiology, and several solutions have been proposed to filter the signal from the noise. These statistical methods are usually referred to as Disease Mapping. In geographical analysis a first goal is to evaluate the statistical significance of the heterogeneity between areas (or points). If the test indicates rejection of the hypothesis of homogeneity the following task is to study the spatial pattern of the disease. The spatial variability of risk is usually decomposed into two terms: a spatially structured (clustering) and a non spatially structured (heterogeneity) one. The heterogeneity term reflects spatial variability due to intrinsic characteristics of the sampling units (e.g. igienic conditions of farms), while the clustering term models the association due to proximity between sampling units, that usually depends on ecological conditions that vary over the study area and that affect in similar way breedings that are close to each other. Hierarchical bayesian models are the main tool to make inference over the clustering and heterogeneity components. The results are based on the marginal posterior distributions of the parameters of the model, that are approximated by Monte Carlo Markov Chain methods. Different models can be defined depending on the terms that are considered, namely a model with only the clustering term, a model with only the heterogeneity term and a model where both are included. Model selection criteria based on a compromise between degree of complexity and goodness of fit are then needed to discriminate among them, because each specification has a different biological meaning. Our aim is to demonstrate that these techniques can be used to study the geographical distribution of a parasite infection. Our analyses are based on data collected in 142 farms of the province of Latina. In each breeding a fixed number of sheeps has been sampled (20) and checked for the presence of C. daubneyi. We have specified a Binomial model for the proportion of infected animals in each breeding. The heterogeneity component is modelled in a standard way, while we have used different prior specifications for the clustering term to show how they affect the results. When we use the usual specification also for clustering, the two models show a completely different spatial pattern of infection, probably because the intrinsic spatial structure of the clustering term tend to bias our inferences. The selection criterion indicates in this case the heterogeneity model as the "best" one. However, if we modify the prior so that a lower degree of spatial interaction is assumed, the clustering model is less complex and its goodness of fit better and it should be preferred.

Animal Husbandry↗
PubMed1991

Conformation of receptor-associated PGI2: an investigation by molecular modeling.

To elucidate the conformation of receptor-associated prostacyclin (PGI2), we first performed structure-activity correlation analysis of over 200 PGI2 analogues and derived from this analysis several crucial features pertaining to structural requirements for PGI2 activity [Ah-lim Tsai and Kenneth K. Wu, Eicosanoids, 2 (1989) 131-143]. These structural features proved to be useful guidelines for selecting 'model molecules' for further investigations by molecular mechanics. By properly selecting four analogues with either rigid or uniquely oriented alpha-side chain structure for geometric fitting, we succeeded in maximally minimizing the degree of freedom of the carboxylate terminus of PGI2. We were able to define the spatial relationship among the four critical functional groups, i.e., C1-COOH, C6a-O, C11-OH and C15-OH. More information is needed, however, to define the geometry of the omega-side chain, particularly for the moiety beyond C15. Nevertheless, results from structure-activity correlation analysis and molecular modeling provide useful information regarding the conformation of receptor-associated PGI2, which assumes an 'elongated' conformation instead of the traditional 'hairpin' structure.

Binding Sites↗
PubMed1992

Reactor paper concerning patient selection and appliance design in intra-oral models.

Patient selection and appliance design are important factors in the development of intra-oral models. Adults, rather than children, are recommended as panelists due to their similarity to children in response to fluoride for caries prevention, greater likelihood of compliance with all study requirements, and greater availability for appointments. It may be desirable for additional factors--including the presence of cariogenic micro-organisms and the characteristics of their saliva with regard to remineralization--to be used in the selection of panelists. Cross-over test designs may be desirable to minimize the impact of these variables as well as of dietary differences. Appliances should be designed to approximate the normal oral conditions to the maximum extent possible. Intra-oral models should be capable not only of detecting the effects of clinically-proven fluoride treatments but also of detecting fluoride dosage responses.

Adult↗
PubMedDate not supplied

Estimating the incidence of dementia from two-phase sampling with non-ignorable missing data.

Two-phase sampling designs have been used in the field of psychiatry to estimate prevalence and incidence of a rare disease such as dementia and Alzheimer's disease. In a longitudinal study on dementia, since the repeated two-phase sampling is conducted several years after the baseline wave, some subjects may die before the follow-up wave, thus their disease status prior to death is missing. There are reasons to suggest that the missing due to death is non-ignorable. Estimation of disease incidence from longitudinal dementia study has to appropriately adjust for data missing by death as well as the sampling design used at each study wave. In this paper we adopt a selection model approach to model the missing data by death and use a likelihood approach to derive incidence estimates. A modified EM algorithm is used to deal with data from sampling selection. The non-parametric jack-knife variance estimator is used to derive variance estimates for the model parameters and the incidence estimates. The proposed approaches are applied to data from the Indianapolis-Ibadan Dementia Study.

Black or African American↗
PubMed2001

A mixed effects model for the analysis of ordinal longitudinal pain data subject to informative drop-out.

We extend the model of Pulkstenis et al. that models binary longitudinal data, subject to informative drop-out through remedication, to the ordinal response case. We present a selection model shared-parameter approach that specifies mixed models for both ordinal response and discrete survival time to remedication. In this fashion, the random parameter present in both models completely characterizes the relationship between response and time to remedication inducing their conditional independence. With a log-log link function for both response and study 'survival', as well as specification of a log-gamma distribution for the random effect, we obtain a closed-form expression for the marginal log-likelihood of response and time to remedication that does not require approximation or numerical integration techniques. A data analysis is performed and simulation results presented which support the consistency of parameter and standard error estimates.

Computer Simulation↗
PubMed1977

Remark on models for amplification of asymmetry.

There exists a number of evolutionary models in which a tiny asymmetry in the size of several subpopulations is amplified. In the present paper a counterexample is given based on equally justifiable starting conditions. The existence of similar models leading to contradictory results clearly demonstrates the highly hypothetic nature and questionable value of conclusions drawn from 'by chance' selection models.

Biological Evolution↗
PubMed1993

Photodegradation of the carbamate insecticide pirimicarb.

In order to study the photoreactivity of pirimicarb (2-(dimethylamino)-5,6-dimethyl-pyrimidin-4-yl-N-dimethylcarbamate ) on plant surfaces, model experiments with organic solvents were performed. Pirimicarb decomposed readily when irradiated (lambda > 280 nm) in the presence of the selected model solvents to form a number of products. Half-lives were in the order isopropanol < cyclohexane < cyclohexene (up to a maximum of 140 min). In the presence of all three solvents photooxidation predominated leading to N-formylpirimicarb and, in a second step, to N-desmethylpirimicarb as well as to further oxygenated products with the carbamate moiety intact. Photolysis in cyclohexene resulted in photomineralisation as the main degradation pathway. Irradiation of pirimicarb in the presence of isopropanol also yielded an addition product of the parent compound with the secondary alcohol.

1-Propanol↗
PubMed2026

Robotic assistance in total hip arthroplasty: a systematic review and meta-analysis of leg length, cup orientation, and early outcomes.

This review examined whether robotic assistance alters postoperative leg-length discrepancy (LLD), acetabular cup orientation, or early hip-specific outcomes relative to conventional total hip arthroplasty (THA). We searched PubMed and Web of Science through May 2026 for comparative English-language reports. Study eligibility, data extraction, and methodological appraisal were undertaken independently by two reviewers. Mean differences (MDs) and 95% confidence intervals (CIs) were calculated in Review Manager 5.4. Model selection was based on the target estimand and anticipated clinical and methodological diversity; leave-one-out and alternative-model sensitivity analyses were undertaken for heterogeneous outcomes. The protocol is registered with PROSPERO (CRD420261454043). The review included seven studies and 968 participants. Compared with conventional THA, robot-assisted THA yielded a smaller postoperative LLD (MD = -2.02, 95% CI -3.46 to -0.58; P = 0.006) and a higher Harris Hip Score (MD = 2.96, 95% CI 1.12 to 4.80; P = 0.002). Mean cup anteversion was lower in the robotic group (MD = -1.52, 95% CI -2.29 to -0.76; P < 0.0001), whereas cup inclination did not differ (MD = -0.71, 95% CI -3.26 to 1.83; P = 0.58). The robotic group also had higher Forgotten Joint Score (MD = 14.68, 95% CI 5.02 to 24.33; P = 0.003) and Oxford Hip Score values (MD = 2.61, 95% CI 0.71 to 4.51; P = 0.007). Robotic assistance was linked to a modest improvement in leg-length restoration and to higher scores on several early functional measures. The limited number of studies, predominance of nonrandomized designs, and marked heterogeneity in some analyses temper the certainty of these findings.

Humans↗
PubMed2003

Sociological influences on antidepressant prescribing.

This study examined how patient characteristics, physician characteristics, the physician's interaction with the health care system, and the physician's interaction with the patient influenced whether patients with a depression diagnosis received an antidepressant prescription and whether they received a SSRI antidepressant, a non-SSRI antidepressant, or both. The 1998 National Ambulatory Medical Care Survey (NAMCS), in the USA, was used for the analysis. Logistic regression was used to examine what characteristics influenced whether a patient with a depression diagnosis received an antidepressant prescription. Next, a multinomial logistic regression model was applied to examine the relative risk of using one type of antidepressant versus another among antidepressant users while correcting for possible sample selections using the Heckman selection model. Sixty-seven percent of patients with a depression diagnosis received an antidepressant. Patients who were seeing providers who were not primary care physicians or psychiatrists, self-paying patients, and patients with neurotic depression were significantly less likely to receive an antidepressant prescription. Patients with depression listed as their first diagnosis were significantly more likely to receive an antidepressant prescription. Patients seeing a psychiatrist were more likely than patients seeing a primary care physician to receive a non-SSRI antidepressant than a SSRI antidepressant. Patients belonging to an HMO that had capitated visits were over four times more likely to receive non-SSRI antidepressants than SSRI antidepressants. Patients with major depression were significantly more likely to receive a non-SSRI antidepressant. Patients with depression as their primary diagnosis and patients who saw psychiatrists were significantly more likely to receive both SSRI and non-SSRI antidepressants rather than just SSRI antidepressants. Patient characteristics, physician characteristics, the physician's interaction with the health care system, and the physician's interaction with the patient all influenced antidepressant prescribing. An especially important finding was that insurance status influenced whether patients received an antidepressant. Health care providers need to take the time to help patients without insurance obtain antidepressant medication if it is needed.

Adolescent↗
PubMed2001

Segregation analyses of 1,476 population-based Australian families affected by prostate cancer.

Segregation analyses aim to detect genetic factors that have a major effect on an individual's risk of disease and to describe them in terms of mode of inheritance, age-specific cumulative risk (penetrance), and allele frequency. We conducted single- and two-locus segregation analyses of data from 1,476 men with prostate cancer diagnosed at age <70 years and ascertained through population registries in Melbourne, Sydney, and Perth, Australia, and from their brothers, fathers, and both maternal and paternal lineal uncles. Estimation and model selection were based on asymptotic likelihood theory and were performed through use of the software MENDEL. All two-locus models gave better fits than did single-locus models, even if lineal uncles were excluded or if we censored data (age and disease status) for relatives at 1992, when prostate-specific-antigen testing started to have a major impact on the incidence of prostate cancer in Australia. Among the genetic models that we considered, the best-fitting ones included a dominantly inherited increased risk that was greater, in multiplicative terms, at younger ages, as well as a recessively inherited or X-linked increased risk that was greater, in multiplicative terms, at older ages. The recessive and X-linked effects were strongly confounded, and it was not possible to fit them together. Penetrance to age 80 years was approximately 70% (95% confidence interval [CI] 57%-85%) for the dominant effect and virtually 100% for the recessive and X-linked effects. Approximately 1/30 (95% CI 1/80-1/12) men would carry the dominant risk, and 1/140 (95% CI 1/220-1/90) would carry the recessive risk or 1/200 (95% CI 1/380-1/100) would carry the X-linked risk. Within discussed limitations, these analyses confirm the genetic heterogeneity, of prostate cancer susceptibility, that is becoming evident from linkage analyses, and they may aid future efforts in gene discovery.

Age of Onset↗
PubMed2004

Covalent DNA display as a novel tool for directed evolution of proteins in vitro.

We present a novel method for the directed evolution of polypeptides, which combines in vitro compartmentalization and covalent DNA display. A library of linear DNA fragments is co-packaged with an in vitro transcription/translation mixture in the compartments of a water-in-oil emulsion. Experimental conditions are adjusted so that, in most cases, one compartment contains one DNA molecule. The DNA fragments encode fusion proteins containing a DNA-methyltransferase (M.Hae III), which can form a covalent bond with a 5-fluorodeoxycytidine base at the extremity of the DNA fragment. The resulting library of DNA-protein fusions is extracted from the emulsion and DNA molecules displaying a protein with desired binding properties are selected from the pool of DNA-protein fusions by affinity panning on target antigens. We applied this methodology in model selection experiments, using specific ligands for the capture of peptides and globular proteins bound to DNA. We observed enrichment factors >1000-fold for selections performed in separate emulsions and up to 150-fold for selections performed using mixtures of DNA molecules. M.Hae III could be fused to small globular proteins (such as calmodulin and fibronectin domains), which are ideally suited for the generation of combinatorial libraries and for the isolation of novel binding specificities.

DNA Adducts↗
PubMed2005

The prion protein gene in humans revisited: lessons from a worldwide resequencing study.

Ample evidence has accumulated showing that different coding variants of the PRNP gene confer differential susceptibility for prion diseases. Here we evaluate the patterns of nucleotide variation in PRNP exon 2, which includes all the protein-coding sequence, by resequencing a worldwide sample of 174 humans for 2378 bp. In line with previous studies, we found two main haplotypes differentiated by nonsynonymous substitution in codon 129. Our analyses reveal the worldwide pattern of variation at the PRNP gene to be inconsistent with neutral expectations, indicating instead an excess of low-frequency variants, a footprint of the action of either positive or purifying selection. A comparison of neutrality test statistics for PRNP with other human genes indicates that the signal of positive selection on PRNP is stronger than expected from a possible confounding genome-wide background signal of population expansion. Two main conclusions arise from our analysis. First, the existence of an ancient, stable, balanced polymorphism that has been claimed in a previous study and related to cannibalism can be rejected and is shown to be due to ascertainment bias. Second, our results are consistent with a complex history of selection including mainly positive selection, even if short local periods of balancing selection (Kuru-like episodes), or even a weak purifying selection model, are consistent with our data.

Amino Acid Substitution↗
PubMed2001

A weighted composite dose-response model for human salmonellosis.

This article describes the development of a weighted composite dose-response model for human salmonellosis. Data from previously reported human challenge studies were categorized into two different groups representing low and moderately virulent/pathogenic Salmonella strains based on a disease end point. Because epidemiological data indicate that some Salmonella strains are particularly pathogenic, and in the absence of human feeding study data for such strains, Shigella dysenteriae was used as a proxy for highly virulent strains. Three single-hit dose-response models were applied to the human feeding study data and evaluated for best fit using maximum likelihood estimation: (1) the exponential (E-1pop), (2) the two-subpopulation exponential (E-2pop), and (3) the Beta-Poisson (BP). Based on the goodness-of-fit test, the E-1pop and BP were the best-fit models for low and moderately virulent/pathogenic Salmonella strains, and the E-2pop and BP models were better for highly virulent/pathogenic strains. Epistemic analysis was conducted by determining the degree of confidence associated with the selected models, which was found to be 50%/50% (E-1pop/BP) for low and moderately pathogenic Salmonella strains, and 9.8%/90.2% (E-2pop/BP) for highly virulent strains. The degree of confidence for each component model and variations in the proportion of strains within each virulence/pathogenicity category were incorporated into the overall composite model. This study describes the influence of variation in strain virulence and host susceptibility on the shape of the population dose-response relationship.

Colony Count, Microbial↗
PubMed2001

A computational model of action selection in the basal ganglia. II. Analysis and simulation of behaviour.

In a companion paper a new functional architecture was proposed for the basal ganglia based on the premise that these brain structures play a central role in behavioural action selection. The current paper quantitatively describes the properties of the model using analysis and simulation. The decomposition of the basal ganglia into selection and control pathways is supported in several ways. First, several elegant features are exposed--capacity scaling, enhanced selectivity and synergistic dopamine modulation--which might be expected to exist in a well designed action selection mechanism. The discovery of these features also lends support to the computational premise of selection that underpins our model. Second, good matches between model globus pallidus external segment output and globus pallidus internal segment and substantia nigra reticulata area output, and neurophysiological data, have been found which are indicative of common architectural features in the model and biological basal ganglia. Third, the behaviour of the model as a signal selection mechanism has parallels with some kinds of action selection observed in animals under various levels of dopaminergic modulation.

Animals↗
PubMed2006

Breast carcinoma specific antibody selection combining phage display and immunomagnetic cell sorting.

To discover new specific antibodies directed against disseminated carcinoma cells in breast cancer patients, a strategy combining single-chain variable fragment (scFv) phage display and immunomagnetic cell sorting was developed. A selection model, in which ErbB2-expressing breast carcinoma SKBR3 cells are spiked into a 50-fold excess of lymphocytes, was setup. Selection conditions, optimized using the previously characterized ErbB2-specific F5 phage scFv, led to an outstanding phage enrichment yield of 25,000 after only one round. This protocol applied to human nai ve and synthetic phage display antibody libraries led to the selection, in only two rounds, of individual scFv clones (43 out of 46 tested) specific for non-epithelial carcinoma antigens expressed on SKBR3 cells. This strategy is fully applicable to metastatic cells in effusions from breast carcinoma patients and shall lead to the discovery of immunotools crucial for novel diagnostic and therapeutic approaches.

Antibodies, Neoplasm↗
PubMedDate not supplied

Monitoring and modeling of the reaction dynamics in the valine/leucine synthesis pathway in Corynebacterium glutamicum.

The intracellular concentrations of the valine and leucine pathway intermediates in a Corynebacterium glutamicum strain were measured during a transient state. The data were obtained by performing a glucose stimulus-response experiment with the use of a rapid sampling device and advanced mass spectrometry. The glucose stimulus resulted in a 3-fold increase in the intracellular pyruvate concentration within less than a second, demonstrating the very fast interactions in metabolic networks. The samples were taken at subsecond intervals for a time period of 25 s. The time courses of the metabolite concentrations formed the experimental basis of a mathematical model simulating the fluxes and concentrations in the valine/leucine pathway. The implementation of a model selection criterion based on the second law of thermodynamics is demonstrated to be essential for the identification of realistic and unique models. Large differences between the enzyme properties determined in vitro and those determined in vivo by the model were observed with the in vivo maximal rates being almost an order of magnitude larger than the in vitro maximal rates. The transamination of ketoisovalerate (KIV) to valine is carried out mainly by the transaminase B enzyme, with the transaminase C enzyme playing a minor role. The availability of the cofactors NADP and NADPH only has modest influence on the flux through the valine pathway, while the influence of NAD and NADH on the flux through the leucine pathway is negligible.

Corynebacterium glutamicum↗
PubMed2001

A structural mode-coupling approach to 15N NMR relaxation in proteins.

The two-body Slowly Relaxing Local Structure (SRLS) model was applied to (15)N NMR spin relaxation in proteins and compared with the commonly used original and extended model-free (MF) approaches. In MF, the dynamic modes are assumed to be decoupled, local ordering at the N-H sites is represented by generalized order parameters, and internal motions are described by effective correlation times. SRLS accounts for dynamical coupling between the global diffusion of the protein and the internal motion of the N-H bond vector. The local ordering associated with the coupling potential and the internal N-H diffusion are tensors with orientations that may be tilted relative to the global diffusion and magnetic frames. SRLS generates spectral density functions that differ from the MF formulas. The MF spectral densities can be regarded as limiting cases of the SRLS spectral density. SRLS-based model-fitting and model-selection schemes similar to the currently used MF-based ones were devised, and a correspondence between analogous SRLS and model-free parameters was established. It was found that experimental NMR data are sensitive to the presence of mixed modes. Our results showed that MF can significantly overestimate order parameters and underestimate local motion correlation times in proteins. The extent of these digressions in the derived microdynamic parameters is estimated in the various parameter ranges, and correlated with the time scale separation between local and global motions. The SRLS-based analysis was tested extensively on (15)N relaxation data from several isotropically tumbling proteins. The results of SRLS-based fitting are illustrated with RNase H from E. coli, a protein extensively studied previously with MF.

Diffusion↗