Elicitation of multivariate prior distributions: A nonparametric Bayesian approach

In the context of Bayesian statistical analysis, elicitation is the process of formulating a prior density f(.) about one or more uncertain quantities to represent a person's knowledge and beliefs. Several different methods of eliciting prior distributions for one unknown parameter have been proposed. However, there are relatively few methods for specifying a multivariate prior distribution and most are just applicable to specific classes of problems and/or based on restrictive conditions, such as independence of variables. Besides, many of these procedures require the elicitation of variances and correlations, and sometimes elicitation of hyperparameters which are difficult for experts to specify in practice. Garthwaite et al. (2005) discuss the different methods proposed in the literature and the difficulties of eliciting multivariate prior distributions. We describe a flexible method of eliciting multivariate prior distributions applicable to a wide class of practical problems. Our approach does not assume a parametric form for the unknown prior density f(.), instead we use nonparametric Bayesian inference, modelling f(.) by a Gaussian process prior distribution. The expert is then asked to specify certain summaries of his/her distribution, such as the mean, mode, marginal quantiles and a small number of joint probabilities. The analyst receives that information, treating it as a data set D with which to update his/her prior beliefs to obtain the posterior distribution for f(.). Theoretical properties of joint and marginal priors are derived and numerical illustrations to demonstrate our approach are given. (C) 2010 Elsevier B.V. All rights reserved.

Propolis Standardized Extract (EPP-AF (R)), an Innovative Chemically and Biologically Reproducible Pharmaceutical Compound for Treating Wounds

The aim of this study was to develop a formulation, containing the propolis standardized extract (EPP-AF (R)), which can assist in the healing of skin lesions. To achieve this objective the antimicrobial activity and chemical composition of the propolis extract was determined. The final product was subjected to in vitro and in vivo pre-clinical evaluation. The broth macrodi-lution method was used to determine the antimicrobial activity of the extracts and formulations against the microorganisms most commonly found in burns, Pseudomonas aeruginosa, Klebsiella pneumoniae, Escherichia coli, Staphylococcus aureus and Staphylococcus epidermidis. Wistar rats with puncture wounded skin were used to evaluate the wound healing properties of propolis. The results of chemical and biological characterization demonstrated the batch-to-batch reproducibility of the standardized extract which is an unprecedented result. The antimicrobial and wound healing activity of the pharmaceutical studied showed the best results when samples contain 3.6% propolis, suggesting that this is the most promising composition.

The in vitro antiviral activity of an aliphatic nitro compound from Heteropteris aphrodisiaca

We investigated the antiviral activity of an aliphatic nitro compound (NC) isolated from Heteropteris aphrodisiaca O. Mach. (Malpighiaceae), a Brazilian medicinal ptant. The NC was tested for its antiviral activity against poliovirus type 1 (PV-1) and bovine herpes virus type 1 (BHV-1) by plaque reduction assay in cell culture. The NC showed a moderate antiviral activity against PV-1 and BHV-1 in HEp-2 celts, and the 50% inhibitory concentration (IC50) were 22.01 mu g/ml (selectivity index (SI) = 2.83) and 21.10 mu g/mi (SI = 2.95), respectivety. At the highest concentration of the drug (40 mu g/ml) a reduction of approximately 80% in plaque assay was observed for both viruses. The treatment of cells or virus prior to infection did not inhibit the replication of virus strains. (C) 2006 Elsevier GmbH. All rights reserved.

Effect of non-steroidal anti-inflammatory drugs (NSAID) on the histamine release induced by compound 48/80 and cardiac anaphylaxis in guinea-pig isolated heart

Histamine release from guinea pig heart treated with compound 48/80 was potentiated by the cyclooxygenase inhibitors indomethacin and piroxicam but not by aspirin or phenylbutazone. This differential effect suggests that the potentiation is not merely due to an inhibition of prostaglandin synthesis. Piroxicam potentiated the histamine release induced by cardiac anaphylaxis whereas indomethacin reduced this effect. The SRS-A antagonist FPL 55712 inhibited histamine release induced by cardiac anaphylaxis, but not that evoked by compound 48/80, and also prevented the potentiation due to indomethacin and piroxicam. In total, these data suggest that the potentiation of histamine release by piroxicam and indomethacin is probably due to a diversion of arachidonic acid metabolism from the cyclooxygenase to the lipoxygenase pathways. The resulting lipoxygenase products may then regulate histamine release, with the secretion due to antigen being more sensitive to such modulation than that evoked by compound 48/80.

Quantum and semiclassical Husimi distributions for a one-dimensional resonant system

We compare exact and semiclassical Husimi distributions for the single eigenstates of a one-dimensional resonant Hamiltonian. We find that both distributions concentrate near the unstable fixed points even when these points are made complex by suitably varying a parameter. © 1992 The American Physical Society.

Linamarin - The toxic compound of cassava

Cassava is a widely grown root crop which accumulates two cyanogenic glucosides, linamarin and lotaustralin. Linamarin accounts for more than 80% of the cassava cyanogenic glucosides. It is a β-glucoside of acetone cyanohydrin and ethyl-methyl-ketone-cyanohydrin. Linamarin β-linkage can only be broken under high pressure, high temperature and use of mineral acids, while its enzymatic break occurs easily. Linamarase, an endogenous cassava enzyme, can break this β-linkage. The enzymatic reaction occurs under optimum conditions at 25°C, at pH 5.5 to 6.0. Linamarin is present in all parts of the cassava plant, being more concentrated on the root and leaves. If the enzyme and substrate are joined, a good detoxification can occur. All the cassava plant species are known to contain cyanide. Toxicity caused by free cyanide (CN-) has already been reported, while toxicity caused by glucoside has not. The lethal dose of CN- is 1 mg/kg of live weight; hence, cassava root classification into toxic and non-toxic depending on the amount of cyanide in the root. Should the cyanide content be high enough to exceed such a dose, the root is regarded as toxic. Values from 15 to 400 ppm (mg CN-/kg of fresh weight) of hydrocyanic acid in cassava roots have been mentioned in the literature. However, more frequent values in the interval 30 to 150 ppm have been observed. Processed cassava food consumed in Brazil is safe in regard to cyanide toxicity.

Inelasticity distributions in high-energy p-nucleus collisions

Inelasticity distributions in high-energy p-nucleus collisions are computed in the framework of the interacting gluon model, with the impact-parameter fluctuation included. A proper account of the peripheral events by this fluctuation has shown to be vital for the overall agreement with several reported data. The energy dependence is found to be weak.

A description of ligand field effects in the Di-μ-Azido-Bis [{Azido(N,N-diethylethylenediamine)}Copper(II)] compound by the simple overlap model

We present a theoretical description of ligand field effects in the di-μ-azido- bis[{azido(N,N-diethylethylenediamine)} copper(II)] compound by the Simple Overlap Model. The ligand field Hamiltonian is expressed in terms of irreducible tensor operators for an assumed D3h site symmetry occupied by the copper ion. The ligand field parameters, calculated from the available structural data, indicate that the copper ion is under the influence of a very strong ligand field. The energy of the d-d absorption band is well reproduced phenomenologically by the model.

Some Extensions of M-Fractions Related to Strong Stieltjes Distributions

The purpose of this paper is to show the symmetric relations that appear between the coefficients of some even and odd extensions of the M-fractions related to a certain kind of symmetric strong Stieltjes distribution.

Nonforward parton distributions of the pion within an effective single instanton approximation

We develop a relativistic quark model for pion structure, which incorporates the nontrivial structure of the vacuum of quantum chromodynamics as modelled by instantons. Pions are bound states of quarks and the strong quark-pion vertex is determined from an instanton induced effective Lagrangian. The interaction of the constituents of the pion with the external electromagnetic field is introduced in gauge invariant form. The parameters of the model, i.e., effective instanton radius and constituent quark mass, are obtained from the vacuum expectation values of the lowest dimensional quark and gluon operators and the low-energy observables of the pion. We apply the formalism to the calculation of the pion form factor by means of the isovector nonforward parton distributions and find agreement with the experimental data. © 2000 Elsevier Science B.V.

Off-Diagonal Quark Distributions in Pions in the Effective Single-Instanton Approximation

Nonperturbative functions that parametrize off-diagonal hadronic matrix elements of the light-cone leading-twist quark operators are considered. These functions are calculated within the proposed relativistic quark model allowing for the nontrivial structure of the QCD vacuum, special attention being given to gauge invariance. Hadrons are treated as bound states of quarks; strong-interaction quark-pion vertices are described by effective interaction Lagrangians generated by instantons. The parameters of the instanton vacuum, such as the effective radius of the instanton and the quark mass, are related to the vacuum expectation values of the quark-gluon operators of the lowest dimension and to low-energy pion observables. © 2000 MAIK Nauka/Interperiodica.

Robust linear mixed models with normal/independent distributions and Bayesian MCMC implementation

Linear mixed effects models have been widely used in analysis of data where responses are clustered around some random effects, so it is not reasonable to assume independence between observations in the same cluster. In most biological applications, it is assumed that the distributions of the random effects and of the residuals are Gaussian. This makes inferences vulnerable to the presence of outliers. Here, linear mixed effects models with normal/independent residual distributions for robust inferences are described. Specific distributions examined include univariate and multivariate versions of the Student-t, the slash and the contaminated normal. A Bayesian framework is adopted and Markov chain Monte Carlo is used to carry out the posterior analysis. The procedures are illustrated using birth weight data on rats in a texicological experiment. Results from the Gaussian and robust models are contrasted, and it is shown how the implementation can be used for outlier detection. The thick-tailed distributions provide an appealing robust alternative to the Gaussian process in linear mixed models, and they are easily implemented using data augmentation and MCMC techniques.