Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “probabilistic modelling”

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 253 records · Page 14Linked to original sources
PubMed2006

Unsupervised analysis of polyphonic music by sparse coding.

We investigate a data-driven approach to the analysis and transcription of polyphonic music, using a probabilistic model which is able to find sparse linear decompositions of a sequence of short-term Fourier spectra. The resulting system represents each input spectrum as a weighted sum of a small number of "atomic" spectra chosen from a larger dictionary; this dictionary is, in turn, learned from the data in such a way as to represent the given training set in an (information theoretically) efficient way. When exposed to examples of polyphonic music, most of the dictionary elements take on the spectral characteristics of individual notes in the music, so that the sparse decomposition can be used to identify the notes in a polyphonic mixture. Our approach differs from other methods of polyphonic analysis based on spectral decomposition by combining all of the following: (a) a formulation in terms of an explicitly given probabilistic model, in which the process estimating which notes are present corresponds naturally with the inference of latent variables in the model; (b) a particularly simple generative model, motivated by very general considerations about efficient coding, that makes very few assumptions about the musical origins of the signals being processed; and (c) the ability to learn a dictionary of atomic spectra (most of which converge to harmonic spectral profiles associated with specific notes) from polyphonic examples alone-no separate training on monophonic examples is required.

Journal Article↗
PubMed2003

Probabilistic analysis indicates discordant gene trees in chloroplast evolution.

Analyses of whole-genome data often reveal that some genes have evolutionary histories that diverge from the majority phylogeny estimated for the entire genome. We present a probabilistic model that deals with heterogeneity among gene trees, implement it via the Gibbs sampler, and apply it to the plastid genome. Plastids and their genomes are transmitted as a single block without recombination, hence homogeneity among gene trees within this genome is expected. Nevertheless, previous work has revealed clear heterogeneity among plastid genes (e.g., Delwiche and Palmer 1996). Other studies, using whole plastid genomes of various algae and land plants, found little additional heterogeneity (Martin et al. 1998; Adachi et al. 2000). We augment the earlier studies by using a data set of 14 taxa: 6 land plants, 2 green algae, a diatom, 2 red algae and a cryptophyte, the cyanelle of the glaucocystophyte Cyanophora, and the blue-green alga Synechocystis as an outgroup. Contrary to the earlier analyses, we cannot find even a single, dominant consensus tree. Therefore, we formulate a probabilistic model that divides the genes into two sets: those that follow the consensus tree and those that have independent gene trees. No particular tree is supported by more than three-fourths of the genes. But the set of genes that follows a certain tree is fairly independent of data processing and the method of analysis. With one possible exception, we find no evidence for collinear or functionally related genes to follow similar trees. The phylogenetic pattern also seems independent of bias in amino acid composition. Among possible explanations for the observed phenomenon, the hypothesis that different genes have different covarion structures is difficult to assess. But gene duplication may be possible through the inverted or direct repeat regions, while horizontal gene transfer seems less likely. In contrast to green algae and land plants, inverted repeat regions in red algae and in Cyanophora show abundant differences among the copies. Thus, genes may get duplicated when they are recruited into the inverted repeat region and one of the two copies may be lost after leaving the inverted repeat region.

Chloroplasts↗
PubMed1997

A Probabilistic Feature Model for Unfolding Tested for Perfect and Imperfect Nestings

Complex stimuli can be represented as sets of features. Coombs' unfolding theory for preferential choice and for comparative similarity judgement is modeled for a feature representation of stimuli. This is called the Additive Random Feature model for Unfolding (ARFU). A special case is obtained if Restle's dissimilarity model applies, which we call ARFU-R. Its implications were tested for verbal and pictorial stimuli which exhibit a nesting or an imperfect nesting of their features. Nearly all of ARFU-R's predictions were satisfied at a 10% level of significance, which establishes the validity of ARFU-R for the stimuli used in a triadic comparisons task. The most salient results were (a) the low level of transitivity of subjects' choices, and (b) the subjects' failure to choose the standard when it was among the alternatives presented to the subject. Because of the nature of the task and the stimuli, these results would appear to apply also to two "real-life" tasks: Diagnosis of a mental illness syndrome and construction of an identikit picture of an offender. Copyright 1997 Academic Press. Copyright 1997 Academic Press

Journal Article↗
PubMed2006

Cost-effectiveness of referrals to high-volume hospitals: an analysis based on a probabilistic Markov model for hip fracture surgeries.

Previous studies suggest that German hospitals that perform a high volume of hip fracture surgeries have a lower mortality rate and shorter length of stay than low-volume hospitals. The goal of this paper was to determine the long-term cost-effectiveness (over 20 years) of referring hip fracture surgeries to high-volume hospitals, based on data from previous studies. From a societal perspective, the cost-effectiveness ratio was 15,530 Euro per QALY (quality-adjusted life year) (95% confidence interval 9,864-33,048 Euro), while total annual costs to the statutory health insurance amounted up to 19.6 million Euro. The referral of hip fracture surgeries to high-volume hospitals is thus likely to be cost-effective in Germany.

Aged↗
PubMed1995

Methods for displaying macromolecular structural uncertainty: application to the globins.

Most molecular graphics programs ignore any uncertainty in the atomic coordinates being displayed. Structures are displayed in terms of perfect points, spheres, and lines with no uncertainty. However, all experimental methods for defining structures, and many methods for predicting and comparing structures, associate uncertainties with each atomic coordinate. We have developed graphical representations that highlight these uncertainties. These representations are encapsulated in a new interactive display program, PROTEAND. PROTEAND represents structural uncertainty in three ways: (1) The traditional way: The program shows a collection of structures as superposed and overlapped stick-figure models. (2) Ellipsoids: At each atom position, the program shows an ellipsoid derived from a three-dimensional Gaussian model of uncertainty. This probabilistic model provides additional information about the relationship between atoms that can be displayed as a correlation matrix. (3) Rigid-body volumes: Using clouds of dots, the program can show the range of rigid-body motion of selected substructures, such as individual alpha helices. We illustrate the utility of these display modalities by the applying PROTEAND to the globin family of proteins, and show that certain types of structural variation are best illustrated with different methods of display.

Animals↗
PubMed2006

Collective responsibility for freeway rear-ending accidents? An application of probabilistic casual models.

Determining whether or not an event was a cause of a road accident often involves determining the truth of a counterfactual conditional, where what happened is compared to what would have happened had the supposed cause been absent. Using structural causal models, Pearl and his associates have recently developed a rigorous method for posing and answering causal questions, and this approach is especially well suited to the reconstruction and analysis of road accidents. Here, we applied these methods to three freeway rear-end collisions. Starting with video recordings of the accidents, trajectory information for a platoon of vehicles involved in and preceding the collision was extracted from the video record, and this information was used to estimate each driver's initial speed, following distance, reaction time, and braking rate. Using Brill's model of rear-end accidents, it was then possible to simulate what would have happened, other things being equal, had certain driver actions been other than they were. In each of the three accidents we found evidence that: (1) short following headways by the colliding drivers were probable causal factors for the collisions, (2) for each collision, at least one driver ahead of the colliding vehicles probably had a reaction time that was longer than his or her following headway, and (3) had that driver's reaction time been equal to his or her following headway, the rear-end collision probably would not have happened.

Accidents, Traffic↗
PubMed2003

3-D structure perceived from dynamic information: a new theory.

Image movement provides one of the most potent two-dimensional cues for depth. From motion cues alone, the brain is capable of deriving a three-dimensional representation of distant objects. For many decades, theoretical and empirical investigations into this ability have interpreted these percepts as faithful copies of the projected 3-D structures. Here we review empirical findings showing that perceived 3-D shape from motion is not veridical and cannot be accounted for by the current models. We present a probabilistic model based on a local analysis of optic flow. Although such a model does not guarantee a correct reconstruction of 3-D shape, it is shown to be consistent with human performance.

Journal Article↗
PubMed2002

Bayesian probabilistic network modeling of remifentanil and propofol interaction on wakeup time after closed-loop controlled anesthesia.

OBJECTIVE: Until now, the knowledge of combining anesthetics to obtain an adequate level of anesthesia and to economize wakeup time has been empirical and difficult to represent in quantitative models. Since there is no reason to expect that the effect of non-opioid and opioid anesthetics can be modeled in a simple linear manner, the use of a new computational approach with Bayesian belief network software is demonstrated. METHODS: A data set from a pharmacodynamic study was used where remifentanil was randomly given in three fixed target concentrations (2, 4, and 8 ng/ml) to 62 subjects. Target concentrations of propofol were controlled according to the closed-loop system feedback of the auditory evoked potential index to render modeling unbiased by the level of anesthesia. Time to open eyes was measured to represent wakeup time after surgery. The NETICA version 1.37 software was used on a personal computer for network building, validation, and prediction. RESULTS: After the learning phase, the network was used to generate a series of random cases whose probability distribution matches that of the compiled network. The sampling algorithms used are precise, so that the frequencies of the simulated cases will exactly approach the probabilities of the network and that of the data learned. The graphical display of the predicted wakeup time shows less variability but a more complex interaction pattern than with the unadjusted original data. CONCLUSIONS: Model building and evaluation with Bayesian networks does not depend on underlying linear relationships. Bayesian relationships represent true features of the represented data sample. Data may be sparse, uncertain, stochastic, or imprecise. Multiple platform software that is easy to use is increasingly available. Bayesian networks promise to be versatile tools for building valid, nonlinear, predictive instruments to further gain insight into the complex interaction of anesthetics.

Anesthesia Recovery Period↗
PubMed2005

A comparison of algorithms for inference and learning in probabilistic graphical models.

Research into methods for reasoning under uncertainty is currently one of the most exciting areas of artificial intelligence, largely because it has recently become possible to record, store, and process large amounts of data. While impressive achievements have been made in pattern classification problems such as handwritten character recognition, face detection, speaker identification, and prediction of gene function, it is even more exciting that researchers are on the verge of introducing systems that can perform large-scale combinatorial analyses of data, decomposing the data into interacting components. For example, computational methods for automatic scene analysis are now emerging in the computer vision community. These methods decompose an input image into its constituent objects, lighting conditions, motion patterns, etc. Two of the main challenges are finding effective representations and models in specific applications and finding efficient algorithms for inference and learning in these models. In this paper, we advocate the use of graph-based probability models and their associated inference and learning algorithms. We review exact techniques and various approximate, computationally efficient techniques, including iterated conditional modes, the expectation maximization (EM) algorithm, Gibbs sampling, the mean field method, variational techniques, structured variational techniques and the sum-product algorithm ("loopy" belief propagation). We describe how each technique can be applied in a vision model of multiple, occluding objects and contrast the behaviors and performances of the techniques using a unifying cost function, free energy.

Algorithms↗
PubMedDate not supplied

Assessing the cost-effectiveness of new pharmaceuticals in epilepsy in adults: the results of a probabilistic decision model.

Epilepsy currently affects more than 400,000 people in the United Kingdom and 2.3 million in the United States. Drug therapy is the mainstay of treatment for patients with epilepsy, but therapies vary widely in their mechanism of action and acquisition cost. This article describes a decision model developed for the National Institute for Clinical Excellence in the United Kingdom. It compares the long-term cost-effectiveness of drugs licensed in adults for use in 3 situations: monotherapy for newly diagnosed patients, monotherapy for refractory patients, and combination therapy for refractory patients. The analysis separately considers the treatment of partial and generalized seizures. The full range of pharmaceutical therapies feasibly used in the UK health system was included in the analysis. The analysis showed that, on the basis of existing evidence, for newly diagnosed patients with partial seizures, carbamazepine and valproate are likely to be the most cost-effective mono-therapies. Carbamazepine is likely to be the most cost-effective 2nd-line monotherapy for refractory patients, and oxcarbazepine would probably be the most cost-effective adjunctive therapy for refractory patients if the willingness to pay for additional health benefits is greater than 18,000 pounds per quality-adjusted life year (QALY). For patients with generalized seizures, valproate is most likely to be cost-effective for newly diagnosed patients. For refractory patients, adjunctive topiramate is more cost-effective than monotherapy alone if the willingness to pay for additional health benefits is greater than 35,000 pounds per QALY. There is, however, considerable uncertainty regarding these results. Some of the methodological features of the study will be of value in designing cost-effectiveness analyses of other therapies for chronic conditions. These include the methods used to deal with the absence of head-to-head trial data and the need to reflect time dependency in Markov transition probabilities.

Adult↗
PubMed2016

A probabilistic generative model for quantification of DNA modifications enables analysis of demethylation pathways.

We present a generative model, Lux, to quantify DNA methylation modifications from any combination of bisulfite sequencing approaches, including reduced, oxidative, TET-assisted, chemical-modification assisted, and methylase-assisted bisulfite sequencing data. Lux models all cytosine modifications (C, 5mC, 5hmC, 5fC, and 5caC) simultaneously together with experimental parameters, including bisulfite conversion and oxidation efficiencies, as well as various chemical labeling and protection steps. We show that Lux improves the quantification and comparison of cytosine modification levels and that Lux can process any oxidized methylcytosine sequencing data sets to quantify all cytosine modifications. Analysis of targeted data from Tet2-knockdown embryonic stem cells and T cells during development demonstrates DNA modification quantification at unprecedented detail, quantifies active demethylation pathways and reveals 5hmC localization in putative regulatory regions.

5-Methylcytosine↗
PubMed2005

Prediction and decoding of retinal ganglion cell responses with a probabilistic spiking model.

Sensory encoding in spiking neurons depends on both the integration of sensory inputs and the intrinsic dynamics and variability of spike generation. We show that the stimulus selectivity, reliability, and timing precision of primate retinal ganglion cell (RGC) light responses can be reproduced accurately with a simple model consisting of a leaky integrate-and-fire spike generator driven by a linearly filtered stimulus, a postspike current, and a Gaussian noise current. We fit model parameters for individual RGCs by maximizing the likelihood of observed spike responses to a stochastic visual stimulus. Although compact, the fitted model predicts the detailed time structure of responses to novel stimuli, accurately capturing the interaction between the spiking history and sensory stimulus selectivity. The model also accounts for the variability in responses to repeated stimuli, even when fit to data from a single (nonrepeating) stimulus sequence. Finally, the model can be used to derive an explicit, maximum-likelihood decoding rule for neural spike trains, thus providing a tool for assessing the limitations that spiking variability imposes on sensory performance.

Action Potentials↗
PubMed2004

A bayesian approach to parameter estimation for a crayfish (Procambarus spp.) bioaccumulation model.

Bioaccumulation models are used to describe chemical uptake and clearances by organisms. Averaged input parameter values are traditionally used and yield point estimates of model outputs. Hence, the uncertainty and variability of model predictions are ignored. Probabilistic modeling approaches, such as Monte Carlo simulation and the Bayesian method, have been recommended by the U.S. Environmental Protection Agency to provide a quantitative description of the degree of uncertainty and/or variability in risk estimates in ecological hazards and human health effects. In this study, a Bayesian analysis was conducted to account for the combined uncertainty and variability of model parameters in a crayfish bioaccumulation model. After a 5-d exposure in the LaBranche Wetlands (LA, USA), crayfish were analyzed for polycyclic aromatic hydrocarbon concentrations and lipid fractions. The posterior distribution of model parameters were derived from the joint posterior parameter distributions using a Markov chain Monte Carlo approach and the experimental data. The results were then used to predict the distribution of chrysene concentration versus time in the crayfish to compare the predicted ranges at the different study sites.

Animals↗