Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Generative models”

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 919 records · Page 51Linked to original sources
PubMed1990

[Evaluation of 196 patients with chronic myeloid leukemia based on a standard prognosis model].

The clinical course of 196 patients with chronic myelocytic leukemia (CML) was studied. Prognostic factors were analyzed using a standard prognostic model. From a univariate analysis of patients with nonblastic Philadelphia chromosome-positive CML, splenomegaly, bone marrow fibrosis, percentage of blasts and promyelocytes in the peripheral blood and LDH activity were shown to be factors with a significant negative influence on survival. However, age and the platelet count did not influence survival. The standard prognostic model, generated with the 4 variables (1) percentage of blasts and promyelocytes, (2) spleen size, (3) platelet count and (4) age did not provide a useful representation of risk status in this heterogenous patient population. However, the addition of further variables (LDH, additional chromosomal aberrations, percentage of basophiles/eosinophiles and percentage of bone marrow blasts) to the standard model allowed a separation into 2 patient groups: one with low and the other with intermediate to high risk. Our data support the general validity of the prognostic model; however, the applicability of the model may be compromised in hematologic centers with a heterogenous CML population due to the selection of high-risk patients. In this situation additional risk factors may have to be added to the prognostic formula.

Biomarkers, Tumor↗
PubMed2000

Attractor dynamics in feedforward neural networks.

We study the probabilistic generative models parameterized by feedforward neural networks. An attractor dynamics for probabilistic inference in these models is derived from a mean field approximation for large, layered sigmoidal networks. Fixed points of the dynamics correspond to solutions of the mean field equations, which relate the statistics of each unit to those of its Markov blanket. We establish global convergence of the dynamics by providing a Lyapunov function and show that the dynamics generate the signals required for unsupervised learning. Our results for feedforward networks provide a counterpart to those of Cohen-Grossberg and Hopfield for symmetric networks.

Bayes Theorem↗
PubMed2003

Comparison of Haseman-Elston regression analyses using single, summary, and longitudinal measures of systolic blood pressure.

To compare different strategies for linkage analyses of longitudinal quantitative trait measures, we applied the "revisited" Haseman-Elston (RHE) regression model (the cross product of centered sib-pair trait values is regressed on expected identical-by-descent allele sharing) to cross-sectional, summary, and repeated measurements of systolic blood pressure (SBP) values in replicate 34, randomly selected from the Genetic Analysis Workshop 13 simulated data. RHE linkage scans were performed without knowledge of the generating model using the following phenotypes derived from untreated SBP measurements: the first, the last, the mean, the ratio of the change between the first and last over time, and the estimated linear regression slope coefficient. Estimates of allele sharing in sibling pairs were obtained from the complete genotype data of Cohorts 1 and 2, but linkage analyses were restricted to the five visits of Cohort 2 siblings. Evidence for linkage was suggestive (p < 0.001) at markers neighboring SBP genes Gb35, Gs10, and Gs12, but weaker signals (p < 0.01) were obtained at markers mapping close to Gb34 and Gs11. Linkage to baseline genes Gb34 and Gb35 was best detected using the first SBP measurement, whereas linkage to slope genes Gs10-12 was best detected using the last or mean SBP value. At markers on chromosomes 13 and 21 displaying strongest linkage signals, marginal RHE-type models including repeated SBP measures were fit to test for overall and time-dependent genetic effects. These analyses assumed independent sib pairs and employed generalized estimating equations (GEE) with a first-order autoregressive working correlation structure to adjust for serial correlation present among repeated observations from the same sibling pair.

Adult Children↗
PubMed1996

A mathematical model for hypothetical force distribution between opposing jaws.

In the present study we tested the hypothesis that the lower jaw was not simply a Class III lever in which bite force was simply related to jaw length. Dimensions considered relevant were measured on 58 (29 male) dry skulls housed in the Dart Collection. These values were then applied to a mathematical model in which were combined analyses of presumed forces in both sagittal and frontal planes. Our results show that calculated forces increase linearly along the model of a dental arch until a point Fmax is reached where calculated forces are maximized. Distal to Fmax there is a gradual drop off of force up to the posterior limit of the schematic dental arch. Point Fmax in models generated for males is at 56.54 mm and females at 53.31 mm from a baseline passing through the centroids of the glenoid fossae.

Biomechanical Phenomena↗
PubMed2002

Second-generation mimics of ganglioside GM1 oligosaccharide: a three-dimensional view of their interactions with bacterial enterotoxins by NMR and computational methods.

As a step to delineate a strategy of ligand design for cholera toxin (CT), NMR studies were performed on several mimics of the GM1 ganglioside oligosaccharide. The conformation of these analogues was investigated first in solution and then upon binding to cholera toxin by transferred nuclear Overhauser effect (TR-NOE) measurements. It was demonstrated that CT selects a conformation similar to the global minima of the free saccharides from the ensemble of presented conformations. No evidence of major conformational distortions was obtained, but one or two of the available conformers of the hydroxyacid side chain appear to be selected in the bound state. The NMR data were interpreted with the aid of computer models, generated and analyzed by using a combination of different approaches (MacroModels' MC/EM and MC/SD, Autodock, and GRID). Analysis of the NMR data supported by computational studies allowed us to interpret the experimental observations and to derive workable models of the ligand:toxin complexes. These models suggest that the higher affinity of the (R)-lactic acid derivative 3 may stem from lipophilic interactions with a hydrophobic area in the toxin binding site located in the vicinity of the sialic acid side chain binding region of the CT:GM1 complex, and formed by the side chain of Ile-58 and Lys-34. Thus, the models obtained have allowed us to make useful design suggestions for the improvement of ligand affinity.

Journal Article↗
PubMed2002

Second-generation mimics of ganglioside GM1 oligosaccharide: a three-dimensional view of their interactions with bacterial enterotoxins by NMR and computational methods.

As a step to delineate a strategy of ligand design for cholera toxin (CT), NMR studies were performed on several mimics of the GM1 ganglioside oligosaccharide. The conformation of these analogues was investigated first in solution and then upon binding to cholera toxin by transferred nuclear Overhauser effect (TR-NOE) measurements. It was demonstrated that CT selects a conformation similar to the global minima of the free saccharides from the ensemble of presented conformations. No evidence of major conformational distortions was obtained, but one or two of the available conformers of the hydroxyacid side chain appear to be selected in the bound state. The NMR data were interpreted with the aid of computer models, generated and analyzed by using a combination of different approaches (MacroModels' MC/EM and MC/SD, Autodock, and GRID). Analysis of the NMR data supported by computational studies allowed us to interpret the experimental observations and to derive workable models of the ligand:toxin complexes. These models suggest that the higher affinity of the (R)-lactic acid derivative 3 may stem from lipophilic interactions with a hydrophobic area in the toxin binding site located in the vicinity of the sialic acid side chain binding region of the CT:GM1 complex, and formed by the side chain of Ile-58 and Lys-34. Thus, the models obtained have allowed us to make useful design suggestions for the improvement of ligand affinity.

Binding Sites↗
PubMed1992

AIDS in Ireland: the reporting delay distribution and the implementation of integral equation models.

This paper deals with two basic aspects concerning the modelling of AIDS incidence in the context of Irish data. We describe initially the adjustment of the number of AIDS cases (Xij) to allow for reporting delays, where a simple form of the likelihood function for the Xij is supported by GLIM. Subsequently, we consider the accessibility of numerical solution (through a NAG routine) of the integral equation models generated by the back-projection method for the adjusted AIDS cases. Results for the Irish data are summarized for various choices of the incidence distribution.

Acquired Immunodeficiency Syndrome↗
PubMed1989

Risk, protective factors, and the prevalence of behavioral and emotional disorders in children and adolescents.

The influence of various risk and protective factors on the presence of one or more behavioral or emotional disorders in 3,294 children and adolescents between the ages of 4 and 16 is examined using data from the Ontario Child Health Study conducted in 1983. Multivariate analyses showed that family problems and parental problems heightened the risk for disorder, whereas being a good student, getting along with others, and participation activities reduced the risk of disorder. Some interaction effects between various factors are also reported. The models generated are compared with others reported in the literature. Future research and social policy implications are discussed.

Achievement↗
PubMed2005

Use of isotopic signatures for mass spectral detection of protein adduction by chemically reactive metabolites of bromobenzene: studies with model proteins.

The cytotoxicity of many small organic compounds often apparently derives from their metabolic activation and covalent binding to cellular proteins. It is therefore of considerable interest to be able to determine, for a given protoxin, which metabolites modify which proteins at which sites. Our laboratory has identified more than 45 target proteins for bromobenzene metabolites in liver by peptide mass mapping after two-dimensional electrophoresis. Through all of this work, we have never observed a bromine-containing peptide. We therefore generated model adducted proteins by carbodiimide coupling of Nalpha-acetyl-Ntau-(p-bromophenyl)-L-histidine (1) and Nalpha-acetyl-Nepsilon-(p-bromophenyl)-L-lysine (2) to bovine pancreatic ribonuclease A. For the adducts, RNase-(1)n and RNase-(2)n, mass spectrometry indicated that n = 0-2 and 0-6, respectively. RNase-(2)n was submitted to in-gel and in-solution digestion with trypsin, and the digests were analyzed by matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) and liquid chromatgraphy electrospray ionization MS (LC/ESI-MS) and tandem MS (MS/MS). Sequence coverages observed ranged from 67% with only three modified lysines observed using in-gel digestion and MALDI-TOF analysis, to 100% coverage with all 10 lysines observed in both modified and unmodified form using in-solution digestion and LC/ESI-MS. In the mass spectra of all modified peptides up to 2000 Da, the bromine isotope pattern was obvious by visual inspection; for peptides up to 3600 Da, the isotopic signature could be recognized by visual comparison to simulated spectra. The presence of Br-containing adducts was confirmed by MS/MS analysis of selected peptides. The selection of peaks for MS/MS analysis was significantly facilitated by visual recognition of the bromine isotope pattern, even at signal-to-noise ratios of 10 (or lower in favorable cases). These results indicate that stable isotope labeling may have considerable potential for detecting and locating protein adducts of reactive metabolites.

Bromobenzenes↗
PubMed2003

Conformational properties of nucleotide-based template-competitive HIV-1 reverse transcriptase inhibitors: analysis of enzyme binding modes.

Nucleotides 2-(4-azidophenacyl)thio-1,N6-etheno-2'-deoxyadenosine 5'-triphosphate 1 and its tetrafluoro analog 2 inhibit HIV-1 reverse transcriptase (RT) competitively relative to template. These template-competitive RT inhibitors (TCRTIs) were analyzed for conformational properties by molecular modeling and NMR analysis. Both inhibitors prefer sugar conformations of C2'-endo/C3'-exo with a high-anti glycosidic bond rotation and +sc/ap phosphate conformation (gamma). The major effect of the etheno group is to favor an extended, fully staggered anti conformation in the N1-C2-S-CH2 psi1 side chain rotation, and NMR analysis detects a long range sugar H4' to side chain phenyl meta-H NOE, a result consistent with this compact structure as an important contributor to the solution structure. The binding model generated places the phenyl side chain in a lipophilic pocket in the template grip region of the RT polymerase domain with the Mg-triphosphate complexed to active site carboxylates. The structures of the TCRTIs are compared with that of the template-competitive DNA polymerase inhibitor 2-(4-azidophenacyl)thio-2'-deoxyadenosine 5'-triphosphate 3, and a theoretical model for selectivity is proposed.

Binding, Competitive↗
PubMed2002

Mechanistic model of myocardial energy metabolism under normal and ischemic conditions.

A moderate reduction in coronary blood flow results in decreased myocardial oxygen consumption, accelerated glycolysis, decreased pyruvate oxidation, and lactate accumulation. To quantitatively understand cardiac metabolism during ischemia, we have developed a mechanistic, mathematical model based on biochemical mass balances and reaction kinetics in cardiac cells. By numerical solution of model equations, computer simulations showed the dynamic responses in glucose, fatty acid, glucose-6-phosphate, glycogen, triglyceride, pyruvate, lactate, acetyl-CoA, and free-CoA as well as CO2, O2, phosphocreatine/creatine, nicotinamide adenine dinucleotide (reduced form)/nicotinamide adenine dinucleotide (oxidized form) (NADH/NAD+), and adenosine diphosphate/adenosine triphosphate (ADP/ATP). When myocardial ischemia was simulated by a 60% reduction in coronary blood flow, the model generated myocardial concentrations, uptakes, and fluxes that were consistent with experimental data from in vivo pig studies. After 60 min of ischemia the concentrations of glycogen, phosphocreatine, and ATP were decreased by 60%, 75%, and 50%, respectively. With the onset of ischemia, myocardial lactate concentration increased and the myocardium switched from net consumer to net producer of lactate. Our model predicted a rapid 13-fold increase in NADH/NAD+, but only a twofold increase in the ratio of acetyl-CoA to free-CoA. These findings are consistent with the concept that pyruvate oxidation is inhibited during ischemia partially by the rise in NADH/NAD+.

Adenosine Diphosphate↗
PubMed1981

Stationary gene frequency distribution in the environment fluctuating between two distinct states.

A general method is given to obtain a stationary distribution in a "stochastic" one-dimensional dynamical system in which an environmental parameter specifying the dynamical system is a stationary Markov process with only two states. By applying this method, the exact stationary gene frequency distribution is obtained for a genic selection model in the environment fluctuating between two distinct states. Several limiting stationary distributions are obtained therefrom, and one of them is shown to coincide with a stationary solution of the diffusion equation heuristically derived by us for more general cases. Discussion is given on the relationship between the diffusion equations obtained by various authors starting from discrete, non-overlapping generation models.

Animals↗
PubMed2001

Constrained modeling of spin-labeled major coat protein mutants from M13 bacteriophage in a phospholipid bilayer.

The family of three-dimensional molecular structures of the major coat protein from the M13 bacteriophage, which was determined in detergent micelles by NMR methods, has been analyzed by constrained geometry optimization in a phospholipid environment. A single-layer solvation shell of dioleoyl phosphatidylcholine lipids was built around the protein, after replacing single residues by cysteines with a covalently attached maleimide spin label. Both the residues substituted and the phospholipid were chosen for comparison with site-directed spin labeling EPR measurements of distance and local mobility made previously on membranous assemblies of the M13 coat protein purified from viable mutants. The main criteria for identifying promising candidate structures, out of the 300 single-residue mutant models generated for the membranous state, were 1) lack of steric conflicts with the phospholipid bilayer, 2) good match of the positions of spin-labeled residues along the membrane normal with EPR measurements, and 3) a good match between the sequence profiles of local rotational freedom and a structural restriction parameter for the spin-labeled residues obtained from the model. A single subclass of structure has been identified that best satisfies these criteria simultaneously. The model presented here is useful for the interpretation of future experimental data on membranous M13 coat protein systems. It is also a good starting point for full-scale molecular dynamics simulations and for the design of further site-specific spectroscopic experiments.

Amino Acid Substitution↗
PubMed1998

Modeling control of eye orientation in three dimensions. I. Role of muscle pulleys in determining saccadic trajectory.

This study evaluates the effects of muscle axis shifts on the performance of a vector velocity-position integrator in the CNS. Earlier models of the oculomotor plant assumed that the muscle axes remained fixed relative to the head as the eye rotated into secondary and tertiary eye positions. Under this assumption, the vector integrator model generates torsional transients as the eye moves from secondary to tertiary positions of fixation. The torsional transient represents an eye movement response to a spatial mismatch between the torque axes that remain fixed in the head and the displacement plane that changes by half the angle of the change in eye orientation. When muscle axis shifts were incorporated into the model, the torque axes were closer to the displacement plane at each eye orientation throughout the trajectory, and torsional transients were reduced dramatically. Their size and dynamics were close to reported data. It was also shown that when the muscle torque axes were rotated by 50% of the eye rotation, there was no torsional transient and Listing's law was perfectly obeyed. When muscle torque axes rotated >50%, torsional transients reversed direction compared with what occurred for muscle axis shifts of <50%. The model indicates that Listing's law is implemented by the oculomotor plant subject to a two-dimensional command signal that is confined to the pitch-yaw plane, having zero torsion. Saccades that bring the eye to orientations outside Listing's plane could easily be corrected by a roll pulse that resets the roll state of the velocity-position integrator to zero. This would be a simple implementation of the corrective controller suggested by Van Opstal and colleagues. The model further indicates that muscle axis shifts together with the torque orientation relationship for tissue surrounding the eye and Newton's laws of motion form a sufficient plant model to explain saccadic trajectories and periods of fixation when driven by a vector command confined to the pitch-yaw plane. This implies that the velocity-position integrator is probably realized as a subtractive feedback vector integrator and not as a quaternion-based integrator that implements kinematic transformations to orient the eye.

Eye Movements↗
PubMed2003

Internal models for visual perception.

Although the extrapolation of past perceptual history into the immediate and distant future is a fundamental phenomenon in everyday life, the underlying processing mechanisms are not well understood. A network model consisting of interacting excitatory and inhibitory cell populations coding for stimulus position is used to study the neuronal population response to a continuously moving stimulus. An adaptation mechanism is proposed that offers the possibility to control and modulate motion-induced extrapolation without changing the spatial interaction structure within the network. Using an occluder paradigm, functional advantages of an internally generated model of a moving stimulus are discussed. It is shown that the integration of such a model in processing leads to a faster and more reliable recognition of the input stream and allows for object permanence following occlusion. The modeling results are discussed in relation to recent experimental findings that show motion-induced extrapolation.

Neural Networks, Computer↗
PubMed2026

Molecular Pathways, Target Landscape, and Translational Models in Heart Failure with Preserved Ejection Fraction.

Heart failure with preserved ejection fraction (HFpEF) is a substantial global health burden and the greatest unmet medical need for cardiovascular diseases. It is marked by pronounced clinical heterogeneity and complex multi-system pathophysiology with limited therapeutic options. Progress in developing effective therapeutics is constrained by the inadequacy of experimental models to fully recapitulate the multifactorial nature of the disease. Recent evidence underscores the significant involvement of inflammatory, oxidative, and mitochondrial pathways in the pathogenesis of HFpEF, with non-coding RNAs and epigenetic regulation serving as crucial modulators and prospective therapeutic targets. This review maps the HFpEF target landscape, while critically assessing the mechanistic contributions, translational fidelity, and limitations of existing in vivo and in vitro models. Further, advances are noted among the in vitro technologies, including human cardiac organoids and engineered heart tissues integrated with high-throughput multi-omics and computational modeling, enabling in-depth examination of HFpEF mechanisms. Finally, we underscore the necessity of integrative, systems-level approaches and multi-marker strategies to enhance translational relevance, improve risk stratification, and accelerate development of mechanism-based therapies. Collectively, this review supports phenotypic-guided and mechanism-informed therapeutic development for HFpEF, and provides a roadmap for next generation model development and therapeutic innovation.

Humans↗
PubMed1994

Neuronal network models of motor generation and control.

Computer simulation of neuronal networks is rapidly becoming accepted as a powerful tool in neuroscience. We illustrate the trends in this field by looking at motor generation and control, with examples from recent modeling studies of different systems, including the spinal swimming rhythm generator of the lamprey.

Animals↗
PubMed1993

Dynamics of reflex cocontraction in hermit crab abdomen: experiments and a systems model.

1. Both stretch and release of the ventral superficial muscles (VSM) in the abdomen of the hermit crab, Pagurus pollicarus, activate the VSM motoneurons in the intact animal and in the isolated abdomen. 2. This reflex was studied by recording intracellularly from muscle fibers innervated by single motoneurons during stretch and release of the VSM. The three motoneurons of the right fourth segment respond to both stretch and release with a phasic burst lasting approximately 250 ms. The burst in the two tonic motoneurons has two components, a short burst lasting 10-20 ms, with a latency from the beginning of stretch of 60-90 ms, and a longer burst of variable length, with a latency of 120 ms. Ramp stretches of different amplitudes and velocities were used to show that the first component is proportional to the absolute value of the second derivative of force and the second component to the absolute value of the first derivative of force. 3. Stretch and release of the VSM also simultaneously evoke phasic bursts in the motoneurons of the dorsal superficial muscles and the VSM circular muscles (functional antagonists of the longitudinal VSM), as well as in contralateral homologues of the same segment and in ipsilateral homologues of the next anterior segment. The effect of this coactivation is to stiffen the abdomen in response to perturbations in any direction. 4. Stretch or release of phasic mechanoreceptors in the VSM evokes this reflex. Isometric electrical stimulation of the isolated muscle also activates them, showing that they are transducing changes in force and suggesting that they operate to increase muscle stiffness by positive feedback. 5. A mathematical systems model of this reflex, composed of two parallel pathways activating the motoneurons, was constructed. The first pathway produces a signal proportional to the absolute value of the second derivative of force, the second pathway a signal proportional to the first derivative of force. The sum of the signals from the two pathways is filtered by an adaptation process, which is followed by a low-pass filter representing muscle activation kinetics. The muscle activation signal is then fed back to multiple muscle force. 6. Simulations using this model generate the phasic bursts to stretch and release as well as reproducing the frequency dependence of this reflex. The predominant action of this reflex is to enhance muscle stiffness.

Abdomen↗