PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Model interpretability”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,279 records · Page 71Linked to original sources
PubMed2007

Assessing mechanical function of the zygomatic region in macaques: validation and sensitivity testing of finite element models.

Crucial to the interpretation of the results of any finite element analysis of a skeletal system is a test of the validity of the results and an assessment of the sensitivity of the model parameters. We have therefore developed finite element models of two crania of Macaca fascicularis and investigated their sensitivity to variations in bone material properties, the zygomatico-temporal suture and the loading regimen applied to the zygomatic arch. Maximum principal strains were validated against data derived from ex vivo strain gauge experiments using non-physiological loads applied to the macaque zygomatic arch. Elastic properties of the zygomatic arch bone and the zygomatico-temporal suture obtained by nanoindentation resulted in a high degree of congruence between experimental and simulated strains. The findings also indicated that the presence of a zygomatico-temporal suture in the model produced strains more similar to experimental values than a completely separated or fused arch. Strains were distinctly higher when the load was applied through the modelled superficial masseter compared with loading an array of nodes on the arch. This study demonstrates the importance of the accurate selection of the material properties involved in predicting strains in a finite element model. Furthermore, our findings strongly highlight the influence of the presence of craniofacial sutures on strains experienced in the face. This has implications when investigating craniofacial growth and masticatory function but should generally be taken into account in functional analyses of the craniofacial system of both extant and extinct species.

Animals↗
PubMed2000

Modeling the bacterial photosynthetic reaction center 3: interpretation of effects of site-directed mutagenesis on the special-pair midpoint potential.

Interpretation of changes in midpoint potential of the "special pair" in bacterial photosynthetic reaction centers caused by site-directed mutagenesis is discussed in terms of a simple tight-binding model which relates them to concomitant variations in spin distribution between the two bacteriochlorophyll molecules of the special pair. Our analysis improves on previous similar ones by Allen and co-workers [Artz, K., Williams, J. C., Allen, J. P., Lendzian, F., Rautter, J., and Lubitz, W. (1997) Proc. Natl. Acad. Sci. U.S.A. 94, 13582; Ivancich, A., Artz, K., Williams, J. C., Allen, J. P., and Mattioli, T. A. (1998) Biochemistry 37, 11812] in that it is both more complete, including electron-phonon coupling, and more accurate. It is applied to analyze data for a series of M160 mutants of Rhodobacter sphaeroides, yielding a value of 0.18+/-0.03 eV for the electronic coupling energy between the highest occupied levels of the two bacteriochlorophylls in the wild-type and a value of the energy offset E(o) between the highest occupied molecular orbitals of the L and M bacteriochlorophylls of 0.14+/-0.03 eV. For a mutant in which the electron hole in the special pair cation is located entirely on the reactive (L) side, a potential of 641+/-30 mV with respect to the normal hydrogen electrode is predicted. This agrees well with the average value ca. 650 mV observed for the heterodimer mutant HL(M202) in which the bacteriochlorophyll on the unreactive M side has been replaced by a bacteriopheophytin, causing extensive charge localization. However, the deduced coupling is found to be very sensitive to small changes in the assumptions used in the model, and various important chemical effects remain to be included.

Bacteriochlorophylls↗
PubMed2006

NOXclass: prediction of protein-protein interaction types.

BACKGROUND: Structural models determined by X-ray crystallography play a central role in understanding protein-protein interactions at the molecular level. Interpretation of these models requires the distinction between non-specific crystal packing contacts and biologically relevant interactions. This has been investigated previously and classification approaches have been proposed. However, less attention has been devoted to distinguishing different types of biological interactions. These interactions are classified as obligate and non-obligate according to the effect of the complex formation on the stability of the protomers. So far no automatic classification methods for distinguishing obligate, non-obligate and crystal packing interactions have been made available. RESULTS: Six interface properties have been investigated on a dataset of 243 protein interactions. The six properties have been combined using a support vector machine algorithm, resulting in NOXclass, a classifier for distinguishing obligate, non-obligate and crystal packing interactions. We achieve an accuracy of 91.8% for the classification of these three types of interactions using a leave-one-out cross-validation procedure. CONCLUSION: NOXclass allows the interpretation and analysis of protein quaternary structures. In particular, it generates testable hypotheses regarding the nature of protein-protein interactions, when experimental results are not available. We expect this server will benefit the users of protein structural models, as well as protein crystallographers and NMR spectroscopists. A web server based on the method and the datasets used in this study are available at http://noxclass.bioinf.mpi-inf.mpg.de/.

Algorithms↗
PubMed2007

Tree-based model checking for logistic regression.

A tree procedure is proposed to check the adequacy of a fitted logistic regression model. The proposed method not only makes natural assessment for the logistic model, but also provides clues to amend its lack-of-fit. The resulting tree-augmented logistic model facilitates a refined model with meaningful interpretation. We demonstrate its use via simulation studies and an application to the Pima Indians diabetes data.

Biomedical Research↗
PubMed2001

Migration of fallout-radionuclides in the soil: effect of non-uniformity of the sorption properties on the activity-depth profiles.

Experimentally observed activity-depth profiles of fallout radionuclides in the soil frequently exhibit a comparatively fast moving tail in soil layers below the peak concentration (tailing). Monte Carlo calculations on the basis of the convection-dispersion model show that this phenomenon can be explained by assuming that either the hydraulic properties of the soil (characterised by the diffusion/dispersion coefficient and pore water velocity) or the sorption properties of the soil (characterised by the distribution coefficient Kd), or both, exhibit a horizontal variability according to a log-normal distribution. Modifications of the activity-depth profile due to a Kd value which decreases linearly with depth were examined by using a random walk approach, based also on the convection-dispersion model. In this case, however, a pronounced tailing effect of the activity-depth profile did not result. Interpretation and realistic modelling of an experimentally observed activity-depth profile which exhibits a tailing effect is thus not unambiguously possible without any additional information on the spatial variability of the hydraulic parameters and, independently, also for the sorption properties.

Absorption↗
PubMed1987

Picosecond time-resolved fluorescence of ribonuclease T1. A pH and substrate analogue binding study.

The tryptophyl fluorescence of ribonuclease T1 decays monoexponentially at pH 5.5, tau = 4.04 ns but on increasing pH, a second short-lived component of 1.5 ns appears with a midpoint between pH 6.5 and 7.0. Both components have the same fluorescence spectrum. Acrylamide quenches both fluorescence components, and the short-lived component is quenched fivefold faster than the predominant long component. Binding of the substrate analogue 2'-guanylic acid at pH 5.5 quenches the fluorescence by 20% and introduces a second decay component, tau = 1.16 ns. Acrylamide quenches both tryptophyl decay components, with similar quenching rates. The fluorescence anisotropy decay of ribonuclease T1 was consistent with a molecule the size of ribonuclease T1 surrounded by a single layer of water at pH 7.4, even though the anisotropy decay at pH 5.5 deviated from Stokes-Einstein behavior. The fluorescence data were interpreted with a model where the tryptophyl residue exists in two conformations, remaining in a hydrophobic pocket. The acrylamide quenching is interpreted with electron transfer theory and suggests that one conformer has the nearest atom approximately 3 A from the protein surface, and the other, approximately 2 A.

Acrylamide↗
PubMed2001

Statistical analysis of failure times in total joint replacement.

Time to revision is an important criterion describing the quality of implants in total joint surgery. Estimates of failure probabilities are required to inform a patient about the risk of suffering a reoperation. Also, regression models are used for comparing different prosthesis designs. Typically, patients dying before a revision are considered as censored for time to prosthesis failure. We argue that this technique is inadequate for estimation of failure probabilities and insufficient for comparison of different designs. We propose a new approach based on a competing risk model to account for concurrent mortality. We describe differences in the estimation of failure probabilities and in the interpretation of regression models for implant failure. Additionally, we introduce a random effects term in the regression model to account for potential dependencies in the failure times of bilaterally treated patients. The new approach is illustrated with fictitious data and data from an observational study conducted at a specialized hospital in Switzerland.

Aged↗
PubMed1996

Deformable models in medical image analysis: a survey.

This article surveys deformable models, a promising and vigorously researched computer-assisted medical image analysis technique. Among model-based techniques, deformable models offer a unique and powerful approach to image analysis that combines geometry, physics and approximation theory. They have proven to be effective in segmenting, matching and tracking anatomic structures by exploiting (bottom-up) constraints derived from the image data together with (top-down) a priori knowledge about the location, size and shape of these structures. Deformable models are capable of accommodating the significant variability of biological structures over time and across different individuals. Furthermore, they support highly intuitive interaction mechanisms that, when necessary, allow medical scientists and practitioners to bring their expertise to bear on the model-based image interpretation task. This article reviews the rapidly expanding body of work on the development and application of deformable models to problems of fundamental importance in medical image analysis, including segmentation, shape representation, matching and motion tracking.

Animals↗
PubMed1998

Quantitative modeling of stochastic systems in molecular biology by using stochastic Petri nets.

An integrated understanding of molecular and developmental biology must consider the large number of molecular species involved and the low concentrations of many species in vivo. Quantitative stochastic models of molecular interaction networks can be expressed as stochastic Petri nets (SPNs), a mathematical formalism developed in computer science. Existing software can be used to define molecular interaction networks as SPNs and solve such models for the probability distributions of molecular species. This approach allows biologists to focus on the content of models and their interpretation, rather than their implementation. The standardized format of SPNs also facilitates the replication, extension, and transfer of models between researchers. A simple chemical system is presented to demonstrate the link between stochastic models of molecular interactions and SPNs. The approach is illustrated with examples of models of genetic and biochemical phenomena where the ULTRASAN package is used to present results from numerical analysis and the outcome of simulations.

Animals↗
PubMedDate not supplied

Evidence of surface diffusion of water molecules on proteins of rabbit lens by 1H NMR relaxation measurements.

In this work, we propose a relaxation model for the interpretation of NMR proton spinlattice and spin-spin relaxation times of mammalian lenses. The framework for this model is based on nuclear magnetic spin-lattice relaxation measurements as a function of temperature at different Larmor frequencies for whole rabbit lenses and fragments of the lens. According to this model, two different dynamic processes of the water molecules determine the relaxation behaviour, namely rotational diffusion and translational surface diffusion. These dynamic processes in conjunction with a two site exchange model give a good explanation of all the measured relaxation data. From the experimental data, we were able to obtain the activation parameters for rotational and translational diffusion of bound lens water. Correlation times of 2.1 x 10(-11) sec and 2.5 x 10(-9) sec and activation energies of 20.5 kJ/mol and 22.5 kJ/mol respectively were found at 308K. At low Larmor frequencies (< or = 100 MHz) the longitudinal relaxation is mainly determined by translational surface diffusion of bound water with a mean square displacement of 1.5 nm, whereas at higher frequencies (> or = 300 MHz), rotational diffusion is the main relaxation mechanism. The spin-spin relaxation is determined by translational diffusion over the whole frequency range and therefore shows only a very small dispersion. By our model it is possible to explain: 1) the strikingly large difference between the T1 value and the T2A and T2B values observed in the lens and 2) the different values of the activation energies measured at different fields for the lens.

Animals↗
PubMed2002

Conformational-relaxation models of single-enzyme kinetics.

Fluorescent spectroscopy experiments with single-enzyme molecules yield a large volume of statistical data that can be analyzed and interpreted using stochastic models of enzyme action. Here, we present two models, each based on the mechanism that an enzyme molecule must pass through a sequence of conformational transformations to complete its catalytic turnover cycle. In the simplest model, only one path leading to the release of product is present. In contrast to this, two different catalytic paths are possible in the second considered model. If a cycle is started from an active state, immediately after the previous product release, it follows a different conformational route and is much shorter. Our numerical investigations show that both models generate non-Markovian molecular statistics. However, their memory landscapes and distributions of cycle times are significantly different. The memory landscape of the double-path model bears strong similarity to the recent experimental data for horseradish peroxidase.

Catalysis↗
PubMed1996

Intention-to-treat analyses for incomplete repeated measures data.

In a randomized longitudinal clinical trial designed to evaluate two or more rival treatments, an intent-to-treat analysis requires inclusion of all randomized patients, regardless of whether they remain on protocol for the duration of the study. We propose a piecewise linear random effects model for analyzing longitudinal data where the multivariate outcome can depend upon time spent on treatment. The model assumes that data are available on a random sample of subjects after treatment is terminated, and allows either a pragmatic or explanatory analysis (as defined by Schwartz and Lellouch, 1967, Journal of Chronic Diseases 20, 637-648). Full maximum likelihood estimation of the model parameters is carried out using widely available statistical software for repeated measures with missing data and for nonparametric survival curve estimation. Data from a national, multicenter pediatric AIDS clinical trial are analyzed to illustrate implementation and interpretation of the model.

Acquired Immunodeficiency Syndrome↗
PubMed1999

Interpreting the Inventory of Interpersonal Problems: subscales based on an interpersonal theory model.

There remains considerable debate about a theoretically interpretable subscale structure for the Inventory of Interpersonal Problems (IIP). In this paper items are extracted from the IIP to form a 40-item shortened version (the IIP-40) comprising eight subscales, each of five items, which conform to the eight octant positions within Birtchnell's interpersonal octagon: a version of interpersonal theory. The inter-item reliability of the subscale structure is found to be acceptable when tested against a sample of 150 pre-assessment for psychotherapy IIP completions within the Centre for the Study of Psychotherapy (CSP), University of Kent. The face validity is established through acceptable inter-rater reliability scores in an experiment using blind raters. Subscale scores are shown for patients within CSP and are found to be significantly different between genders in the octant positions of Upper Close and Neutral Distant. High scores in the octant position of Upper Distant is found to be a significant predictor of therapeutic drop-out. The results for gender and therapeutic engagement are consistent with other published work in the interpersonal theory field.

Adult↗
PubMedDate not supplied

Dispersion models and longitudinal data analysis.

Dispersion models provide a flexible class of non-normal distributions with many potential applications in biostatistics, accommodating a wide range of continuous, discrete and mixed data. Starting with Liang and Zeger's generalized estimating equation method, we review some recent applications of dispersion models in longitudinal data analysis, including state space models based on the Tweedie class of exponential dispersion models. In medical applications the latent process of a state space model may often be interpreted as an unobserved potential morbidity process, which is modelled as a function of time varying covariates. By allowing a multivariate response vector of 'symptoms', the model integrates several response variables of mixed types into a single model. For growth curve models, the latent process reflects the 'true' growth.

Air Pollutants↗
PubMed1979

Mlab--a mathematical modeling tool.

An interactive interpreter called Mlab is described. One uses Mlab by typing commands. In this sense, Mlab is a programming language. It has various mathematical and graphical facilities which make it a useful tool for mathematical modeling. The curve fitting capabilities of Mlab are augmented with differential-equation-handling and matrix-manipulation capabilities which provide a powerful and civilized facility for curve fitting. Many people are engaged in this activity, and, in general, they use programs which are neither sufficiently general nor easy to use. (Some conventional programming is usually required, for example.) Mlab purports to be easier than alternate approaches. The nature of Mlab is discussed with accompanying examples. The main example is the use of curve fitting to determine molecular weight from ultracentrifuge data. This example was chosen because it exhibits a special feature of Mlab, namely the root operator, which appears in the definition of the model function.

Computers↗
PubMed1980

The effect of integral conditions in certain equations modelling epidemics and population growth.

Models of epidemics that lead to delay differential equations often have subsidiary integral conditions that are imposed by the interpretation of these models. The neglect of these conditions may lead to solutions that behave in a radically different manner from solutions restricted to obey them. Examples are given of such behavior, including cases where periodic solutions may occur off the natural set defined by these conditions but not on it. A complete stability analysis is also given of a new model of a disease propagated by a vector where these integral conditions play an important role.

Disease Outbreaks↗
PubMed1995

Brownian dynamics simulation of probe diffusion in DNA: effects of probe size, charge and DNA concentration.

We have used Brownian dynamics simulation to study probe diffusion in solutions of short chain DNA using our previously developed simulation algorithm. We have examined the effect of probe size, charge, and DNA concentration on the probe diffusion coefficient, with the aim of gaining insight into the diffusion of proteins in a concentrated DNA environment. In these simulations, DNA was modeled as a worm-like chain of hydrodynamically equivalent spherical frictional elements while probe particles were modeled as spheres of given charge and hydrodynamic radius. The simulations allowed for both short range Lennard-Jones interactions and long ranged electrostatic interactions between charged particles. For uncharged systems, we find that the effects of probe size and DNA concentration on the probe diffusion coefficient are consistent with excluded volume models and we interpret our results in terms of both empirical scaling laws and the predictions of scaled particle theory. For charged systems, we observe that the effects of probe size and charge are most pronounced for the smallest probes and interpret the results in terms of the probe charge density. For an ionic strength of 0.1 M we find that, below a critical probe surface charge density, the probe diffusion coefficient is largely independent of probe charge and only weakly dependent on the DNA charge. These effects are discussed in terms of the interactions between the probe and the DNA matrix and are interpreted in terms of both the underlying physics of transport in concentrated solutions and the assumptions of the simulation model.

Journal Article↗