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 1,405 records · Page 78Linked to original sources

PubMedDate not supplied

In silico ADME/Tox: why models fail.

By way of example, we discuss the apparent 'failure' of in silico ADME/Tox models and attempt to understand the causes. Often, the interpretation of the success of models lies in their use and the expectations of the user. Other times, models are, in fact, of little value. Disappointing results can be linked to the key aspects of the model and modeling procedure, many of these related to the original data and its interpretation. We make recommendations to providers of models regarding the development, description, and use of models as well as the data and information that are important to understanding a model's quality and scope of use.

Carrier Proteins↗

PubMed1986

Carboxylate groups in bovine somatotropin involved in growth promoting activity. Theoretical models based upon individual kinetic constants to interpret the activity decay after chemical modification.

Modification of approximately one fifth of the carboxylate groups in bovine somatotropin with a water soluble carbodiimide caused loss of growth promoting potency pointing to the existence of residues related to the hormonal activity among those belonging to a fast reacting set. A sigmoidal curve was obtained whether the inactivation process was referred to reaction time or degree of modification. Isoelectrofocusing of derivatives released the native hormone from responsibility for the biological potency exerted by preparations with 1.5-2.6 modified carboxylate groups. Examination of the individual reaction kinetics of the 11 fast reacting residues, in turn, excluded the possibility of the sigmoidal character of the inactivation curve being caused by a nonexponential disappearance of essential residues, as a possible consequence of the chemical modification of others. According to synthetic models, the experimental curve may be the consequence of the effect of cumulative modification of 2 or 3 out of a set of 3 to 8 relevant residues.

Carboxypeptidases↗

PubMed1988

Substance P antagonists and analgesia: a review of the hypothesis.

This review focused entirely on the hypothesis that a substance P/neurokinin antagonist should have, and the experimental evidence examining whether they do have, analgesic/antinociceptive properties. Such a hypothesis is reasonable considering the wealth of evidence implicating substance P in the nociceptive process and the demonstration that antibodies to substance P produce or potentiate antinociception. However, despite the availability of several putative antagonists, their pharmacological purity, specificity and selectivity are questionable. Thus, the investigator may not have, as yet, the appropriate tool drug with which to work. Much of the information concerning these points is generated utilizing in vitro (referring to isolated tissue preparations) bioassay tests which may not adequately reflect nor predict their pharmacology in the CNS. Differences in species responsiveness further complicate experimental design and interpretation. Apart from these factors, the choice of test or tests becomes an important consideration. What test, if any, adequately and appropriately reflects the endogenous physiological activity of substance P in nociception and predicts clinically useful activity of an antagonist? Several different models have been described and I have emphasized that conclusions based on a single model should be interpreted with caution. If the ultimate intent of the study is to further define the role of substance P in nociception, then most of the models discussed are adequate. However, if the intent is to demonstrate that a substance P/neurokinin antagonist should have therapeutically useful analgesic activity, it is incumbant on the investigator to demonstrate that, in their model, substance P release is a primary event, the resultant analgesia correlates to the occupancy of the neurokinin receptor by antagonist (ultimately important for all conclusions) and that the model adequately reflects activity of known analgesics in clinical use (validation of the model). In conclusion, given the complexities and contradictions of existing information, the hypothesis that a substance P/neurokinin antagonist should have analgesic/antinociceptive properties remains to be proven.

Amino Acid Sequence↗

PubMed1985

A model of growth with first-order birth and death rates.

This paper describes the derivation, behavior, and application of a model of growth based on elementary kinetic considerations. The model is based on an abbreviated version of the cell cycle in which first-order kinetics govern both production in a growing fraction and loss from the resting fraction of a cell population. Transition between the resting and growing fraction is assumed. The model is derived in the form of a first-order ordinary differential equation with a simple general form, but no closed form solution. Model behavior is examined analytically at equilibrium, in a limiting case, and qualitatively by numerical integration. Because of the simplicity of the underlying assumptions, the model parameters have direct biological interpretations and simple units. Examples of model fitting to data are given, including a direct comparison to the Gompertz growth law.

Animals↗

PubMed1986

Evaluation of a mathematical model analysing the relation between intradental nerve impulse activity and perceived pain in man.

In this investigation the usefulness and accuracy of the parameters of a mathematical model for the analysis of the effectiveness of different pain relieving procedures on pulpal pain were studied. The investigation was carried out using a previously developed mathematical/biological model on data from original subject recordings of Intradental Nerve Impulse Activity (INA) and pain estimations. Computer simulated INA and pain estimation curves were also used to enable calculation of the mathematical and biological conditions that are essential for the actual mathematical/biological model, and to facilitate the interpretation of the parameters of the model. It was shown by means of both real and simulated data that the mathematical model is well suited for the analysis of the effectiveness of some pain relieving procedures on pulpal pain. It was also shown that, by means of three new variables, the model could be made even more accurate and useful for this application.

Computer Simulation↗

PubMed1992

E/M dips. Evidence for lipids regularly distributed into hexagonal super-lattices in pyrene-PC/DMPC binary mixtures at specific concentrations.

We have examined the effect of 1-palmitoyl-2-(10-pyrenyl)decanoyl-sn-glycerol-3-phosphatidylcholine (Pyr-PC) concentration on the ratio of excimer fluorescence to monomer fluorescence (E/M) in L-alpha-dimyristoylphosphatidylcholine (DMPC) multilamellar vesicles at 30 degrees C, with special attention focussed on the smoothness of the curve. We observed a series of dips, in addition to kinks, in the plot of E/M versus the mole fraction of Pyr-PC (XPyrPC). The observation of dips is a new finding, perhaps unique for Pyr-PC in DMPC since only kinks were observed for Pyr-PC in L-alpha-dipalmitoylphosphatidylcholine (DPPC) and in egg yolk phosphatidylcholine (egg-PC) (Somerharju et al., 1985. Biochemistry. 24: 2773-2781). The dips/kinks observed here are distributed according to a well defined pattern reflecting a lateral order in the membrane, and distributed symmetrically with respect to 50 mol% Pyr-PC. Some of the dips appear at specific concentrations (YPyrPC) according to the hexagonal super-lattice model proposed by Virtanen et al. (1988. J. Mol. Electr. 4: 233-236). However, the observations of dips at XPyrPC > 66.7 mol% and the kink at 33.3 mol% cannot be interpreted by the model of Virtanen et al. (1988). These surprising results can be understood by virtue of an extended hexagonal super-lattice model, in which we have proposed that if the pyrene-containing acyl chains are regularly distributed as a hexagonal super-lattice in the DMPC matrix at a specific concentration YPyrPC, then the acyl chains of DMPC can form a regularly distributed hexagonal super-lattice in the membrane at a critical concentration (1-YPyrPC). The excellent agreement between the calculated and the observed dip/kink positions, except for the dip at 74 mol% and the kink at 40 mol%, provides most compelling evidence that lipids are regularly distributed into hexagonal super-lattices in Pyr-PC/DMPC mixtures at specific concentrations. The physical nature of the dips not only gives us a better understanding of lipid lateral organization in membranes but also will lead to new theoretical considerations and experimental designs for exploring the relationship between lipid regular distribution and membrane functions.

1,2-Dipalmitoylphosphatidylcholine↗

PubMed1974

Uptake and release of ferritin iron. Surface effects and exchange within the crystalline core.

The uptake and subsequent release of iron by apoferritin and ferritin was studied by using labelled iron ((59)Fe). The experimental results are consistent with predictions arising from a model system developed in the interpretation of previous experiments. In this model, uptake and release of ferritin iron is controlled by the available surface area of the small crystalline particles of hydrous ferric oxide found within the ferritin molecule. Evidence is also presented for the exchange of Fe(3+) ions among the various cation sites within these crystallites.

Acetates↗

PubMed1999

The "thermolabile" variant of methylenetetrahydrofolate reductase and neural tube defects: An evaluation of genetic risk and the relative importance of the genotypes of the embryo and the mother.

Recent reports have implicated the "thermolabile" (T) variant of methylenetetrahydrofolate reductase (MTHFR) in the causation of folate-dependent neural tube defects (NTDs). We report herein the largest genetic study of NTD cases (n=271) and families (n=218) to date, establishing that, in Ireland, the "TT" genotype is found in 18.8% of cases versus 8.3% of controls (odds ratio 2.57; confidence interval [CI] 1.48-4.45; P=.0005). The maternal and paternal TT genotypes have intermediate frequencies of 13.8% and 11.9%, respectively, indicating that the predominant MTHFR-related genetic effect acts via the TT genotype of the developing embryo. Analysis of the 218 family triads of mother, father, and affected child with log-linear models supports this interpretation, providing significant evidence that the case TT genotype is associated with NTDs (P=.02) but no evidence of a maternal TT genotypic effect (P=. 83). The log-linear model predicted that the risk of NTDs conferred by the case TT genotype is 1.61 (CI 1.06-2.46), consistent with the paramount importance of the case TT genotype in determining risk. There is no compelling evidence for more than a modest additional risk conferred by a maternal TT genotype. These results favor a biological model of MTHFR-related NTD pathogenesis in which suboptimal maternal folate status imposes biochemical stress on the developing embryo, a stress it is ill-equipped to tolerate if it has a TT genotype.

Alleles↗

PubMed1993

Onset of the nondepolarizing neuromuscular block in humans: quantitative aspects.

The purpose of our study was a unified description of the time course for the onset of the neuromuscular block produced by different muscle relaxants after bolus intravenous injections. The environment of the receptors on the motor end plate was assumed to be a part of the interstitial space of the muscle. A unified consideration of four muscle relaxants was accomplished by expressing the concentrations of each relaxant in the interstitial space as multiples of the relaxant's dissociation constant. A flow-limited model was developed to describe the time course of the relaxant's concentration in the interstitial space as a function of its plasma concentration, the plasma flow to the muscle, and the volume of the interstitial space. The results show that those relaxants whose plasma concentrations decrease rapidly achieve an earlier (4-8 min), but relatively lower, peak concentration in the interstitial space. The relaxants with more sustained plasma concentrations reach the peak concentration later, 9-16 min after the bolus intravenous injection. The model allows the interpretation of several observations encountered with the clinical use of the muscle relaxants.

Humans↗

PubMed1992

The findings--diagnosis continuum: implications for image descriptions and clinical databases.

As part of the Unified Medical Language System (UMLS) project, we have been exploring the use of semantic net representation to build a medical ontology that can adapt to the needs and perspective of differing kinds of users with varying purposes. A principal objective is to facilitate indexing and retrieval of objects in a variety of target databases, using their own source vocabularies, while maintaining the representation of concepts to which these source vocabularies refer in a single consistent form, so that retrievals that span resource types can be accommodated. In addition, a particular area of deficiency of the existing UMLS Metathesaurus is that of clinical findings, a part of the problem being the multiple alternative views and granularity levels at which clinical findings are described in different target databases. The problem is particularly obvious when one examines the way in which image findings are described, which may be at a purely perceptual level, or at varying levels of aggregation into higher level observations or interpretations. We have developed a recursive model for representing observations and interpretations in a semantic net along a continuum of degree of aggregation, that appears to lend itself well to adaptation to varying perspectives.

Diagnosis, Computer-Assisted↗

PubMed1984

Response of sensitive human ataxia and resistant T-1 cell lines to accelerated heavy ions.

We have studied the radiation dose responses of two human fibroblast lines: cells from a patient with Ataxia telangiectasia (AT-2SF) and an established line of human T-1 cells. Aerobic and hypoxic 225 kVp X-ray survival curves were used as controls to the heavy ion exposures. Nearly monoenergetic accelerated neon and argon ions were used at the Berkeley Bevalac with various residual range values. The LET of the particles varied from 30 keV microns-1 to over 1,000 keV microns-1. All Ataxia survival curves were exponential functions of the dose. Their radiosensitivity reached peak values at 100-200 keV microns-1. Human T-1 cells have effective sublethal damage repair as has been evidenced by split dose experiments, and they are much more resistant to low LET than to high LET radiation. At high LET their radiosensitivity approached that of the Ataxia cells. The repair-misrepair model has been used to interpret these results. According to this model, the molecular repair processes culminate either in eurepair or in misrepair. We have obtained mathematical expressions that describe the cross sections and inactivation coefficients for both human cell lines as a function of the LET and the type of particle used. We assume that the lesions induced in T-1 and Ataxia cells are qualitatively similar and that each cell line attempts to repair these lesions. The result in most irradiated Ataxia cells, however, is either lethal misrepair or incomplete repair leading to cell death. T-1 cells have efficient repair mechanisms at low LET, and the repair-misrepair model suggests that at high LET the T-1 cells can still efficiently repair individual lesions, but that as the lesions become closely spaced along the tracks, the probability of misrepair increases.

Ataxia Telangiectasia↗

PubMed1996

Modeling Unconscious Gender Bias in Fame Judgments: Finding the Proper Branch of the Correct (Multinomial) Tree

In the preceding article, Buchner and Wippich used a guessing-corrected, multinomial process-dissociation analysis to test whether a gender bias in fame judgments reported by Banaji and Greenwald (Journal of Personality and Social Psychology, 1995, 68, 181-198) was unconscious. In their two experiments, Buchner and Wippich found no evidence for unconscious mediation of this gender bias. Their conclusion can be questioned by noting that (a) the gender difference in familiarity of previously seen names that Buchner and Wippich modeled was different from the gender difference in criterion for fame judgments reported by Banaji and Greenwald, (b) the assumptions of Buchner and Wippich's multinomial model excluded processes that are plausibly involved in the fame judgment task, and (c) the constructs of Buchner and Wippich's model that corresponded most closely to Banaji and Greenwald's gender-bias interpretation were formulated so as to preclude the possibility of modeling that interpretation. Perhaps a more complex multinomial model can model the Banaji and Greenwald interpretation.

Journal Article↗

PubMed2026

Review: The African turquoise killifish as a model for the integrative physiology of vertebrate aging.

With increasing emphasis on extending healthy lifespan, aging research requires vertebrate models that permit efficient mechanistic investigation and intervention testing within practical time and cost constraints. The African turquoise killifish (Nothobranchius furzeri) has attracted growing attention because it combines an exceptionally short life cycle with an intact vertebrate physiological context and an expanding genetic toolkit, enabling relatively rapid evaluation of candidate aging interventions and mechanistic analysis across molecular, tissue, and organismal levels. This review assesses N. furzeri from an integrative-physiology perspective, focusing on germline-soma interactions, gut microbiota-host crosstalk, nutrient sensing and metabolic remodeling, temperature responsiveness, and AMPK-mTOR-linked programs. It also examines expanding genome-engineering and reporter approaches that support mechanistic and tissue-resolved investigation of these physiological processes. Building on recent reviews of killifish biology, disease modeling, regeneration, and the hallmarks of aging, we synthesize evidence across major intervention domains, distinguish established phenotypic effects from incompletely resolved mechanisms, and highlight functional endpoints, methodological standardization, and the appropriate interpretation of the model's translational relevance. Together, these features position N. furzeri as a strategically useful vertebrate platform for rapid mechanistic testing, intervention evaluation, and prioritization of aging-related pathways. Future progress will require improved methodological standardization, tissue-resolved causal studies, and question-driven cross-species validation where appropriate.

Animals↗

PubMed1981

Epidemiologic detection of low dose effects on the developing fetus.

Evaluations of the health effects of exposures in the workplace and environment have broadened to include effects on reproduction, as well as on the development of cancer. Models to assess risks associated with varying doses of exposure rest almost entirely on data about cancer. In this paper we discuss some distinctive features of reproduction which bear on the interpretation of such models, when applied to reproduction, rather than carcinogenesis. Dose-response curves describe the relationship between two exposures (smoking and alcohol drinking) and two outcomes (spontaneous abortion and birthweight) are used to illustrate some of the questions which arise in attempting to determine a "safe" level of exposure.

Abortion, Spontaneous↗

PubMed2001

Mechanism of fast protein folding.

An explosion of in vitro experimental data on the folding of proteins has revealed many examples of folding in the millisecond or faster timescale, often occurring in the absence of stable intermediate states. We review experimental methods for measuring fast protein folding kinetics, and then discuss various analytical models used to interpret these data. Finally, we classify general mechanisms that have been proposed to explain fast protein folding into two catagories, heterogeneous and homogeneous, reflecting the nature of the transition state. One heterogeneous mechanism, the diffusion-collision mechanism, can be used to interpret experimental data for a number of proteins.

Electron Spin Resonance Spectroscopy↗

PubMed1990

Interconversion of metarhodopsins I and II: a branched photointermediate decay model.

Flash photolysis experiments designed to monitor the establishment of the metarhodopsin I to metarhodopsin II equilibrium are interpreted according to a branched model in which two spectrally indistinguishable but kinetically distinguishable forms of metarhodopsin II are postulated to exist in equilibrium with a common pool of metarhodopsin I. This interpretation arises from the consistent requirement for at least three exponentials for a valid description of the observed growth of absorbance at 380 nm following bleaching of bovine rhodopsin in rod outer segment disk membranes. Analysis of the 380-nm transient absorbance data permitted direct determination of the five physically interpretable individual rate constants of the model. This analysis represents a more explicit interpretation of kinetic data than that employed in earlier experiments of this kind, which involved estimating only apparent rates and apparent amplitudes of discrete multiexponential functions. The 380-nm absorbance contributions of all relevant species contributing to the observed dynamic absorbance change were accounted for simultaneously during nonlinear least-squares estimation of the model rate parameters. Analysis of deconvoluted equilibrium spectra acquired from samples identical with those used in the kinetics experiments confirmed the metarhodopsin I-metarhodopsin II equilibrium constants, Keq, derived from the dynamic analyses. It is shown that Keq varies from 1.28 at 10 degrees C to 7.3 at 37 degrees C and that approximately 90% of the metarhodopsin II present is in the form of metarhodopsin IIslow over the temperature range 10-37 degrees C. A physical interpretation of this decay model is discussed in the context of a distribution of metarhodopsin II structural and energetic states.

Cell Membrane↗

PubMed1993

Structure of clathrin-coated vesicles from small-angle scattering experiments.

Previously published small-angle neutron and X-ray scattering data from coated vesicles, reassembled coats, and stripped vesicles have been analyzed in terms of one common model. The neutron data sets include contrast variation measurements at three different D2O solvent concentrations. The model used for interpreting the data has spherical symmetry and explicitly takes into account polydispersity, which is described by a Gaussian distribution. A constant thickness of the clathrin coats is assumed. The fitting of the model shows that the coated vesicles consist of a low-density outer protein shell (clathrin) and a central protein shell (accessory polypeptides and receptors) of approximately six times higher density. For the X-ray scattering and neutron contrast variation data, the polydispersity of the samples is of the order of 90 A (full-width-at-half-maximum value) and the average outer radius is approximately 400 A. The inner high-density shell has inner and outer radii of 115 and 190 A, respectively. A simultaneous fit to the three neutron contrast variation data sets identifies the lipid membrane with a thickness of 40 A and an outer radius of 196 A. Thus, the membrane and the high-density protein shell overlap in space, which shows that the lipid membrane contains protein. The molecular mass of the average particle is 27 x 10(6) Da. The coated vesicles consist, on average, of approximately 85% protein and 15% lipids. About 40% of the protein mass is situated in the central high-density shell, which gives a large amount of protein in the lipid membrane. The densities of the central shell and the lipid membrane show that the hydration is small in the central region. A comparison of the total mass, the mass distribution, and the structure of the average-size particles with the barrel structure shows that the accessory polypeptides are incorporated in the lipid membrane. The results from the neutron data for the reassembled coats show that the structure of these particles is very similar to the structure of the native coats. The main difference is a higher density of the central protein shell, which shows that the membrane is replaced by protein in the reassembled coats.

Clathrin↗

PubMed2026

Machine learning for population-level risk prediction of future cholangiocarcinoma.

BACKGROUND: The poor prognosis of cholangiocarcinoma (CCA) is largely driven by rapid, asymptomatic disease progression, which usually results in a late diagnosis in the absence of established screening strategies. An early, cost-effective, and universally applicable risk assessment strategy would therefore be valuable. METHODS: We developed machine learning (ML) models on prospective, multimodal data from 487,495 UK Biobank (UKB) participants, of whom 649 developed CCA during follow-up. Data from England (80%) were utilised for ML development via five-fold cross-validation, and then all models were tested on withheld data from Scotland, Wales, and Newcastle (20%). Iterative ablation studies reduced inputs from >150 features across demographic data, lifestyle, health records, blood parameters, genomics, and metabolomics to models built on five and ten routinely available clinical parameters. These were externally validated in the Penn Medicine Biobank (PMBB; n = 2638; 28 CCA), All of Us Research Program (AOU; n = 330,433; 362 CCA), Japan Medical Data Centre Claims Database (JMDC; n = 8,425,522; 723 CCA) and TriNetX (n = 728,886; 1592 CCA). FINDINGS: We show that ML models integrating biliary-disease associated health records and Gamma glutamyltransferase can stratify risk of future CCA. Evaluation on the UKB test set as well as three independent cohorts revealed robust performance and generalisability across ethnicities. We achieved AUROCs of 0.71 [95% CI: 0.703-0.711], 0.77 [95% CI: 0.764-0.778 ], 0.796 [95% CI: 0.795-0.798] and 0.8 [95% CI: 0.794-0.805] for UKB, PMBB, AOU, and JMDC respectively, with respective AUPRCs of 0.014 [95% CI: 0.009-0.018], 0.042 [95% CI: 0.037-0.048], 0.038 [95% CI: 0.033-0.042] and 0.001 [95% CI: 0.001-0.001]. In AOU, application of the Youden J-optimised threshold yielded a number needed to screen of 79. Separate models for intra- and extrahepatic CCA did not improve performance. In line with the pathophysiology, performance declined for longer intervals between assessment and event. A group-level analysis in the TriNetX cohort revealed hazard ratios of up to 82.5 [95% CI: 26.4-257.96]. We provide extensive interpretability results and release all source codes used to develop the presented models. INTERPRETATION: We provide a comprehensive framework for early CCA risk stratification in the general population, identifying key predictors, and demonstrating the potential of data-driven models in personalised screening for hepatobiliary cancer. FUNDING: German Cancer Aid (grant #70115730), Junior Principal Investigator Fellowship programme of RWTH Aachen Excellence strategy.

Humans↗