Semiparametric Bayesian measurement error modeling

This work presents a Bayesian semiparametric approach for dealing with regression models where the covariate is measured with error. Given that (1) the error normality assumption is very restrictive, and (2) assuming a specific elliptical distribution for errors (Student-t for example), may be somewhat presumptuous; there is need for more flexible methods, in terms of assuming only symmetry of errors (admitting unknown kurtosis). In this sense, the main advantage of this extended Bayesian approach is the possibility of considering generalizations of the elliptical family of models by using Dirichlet process priors in dependent and independent situations. Conditional posterior distributions are implemented, allowing the use of Markov Chain Monte Carlo (MCMC), to generate the posterior distributions. An interesting result shown is that the Dirichlet process prior is not updated in the case of the dependent elliptical model. Furthermore, an analysis of a real data set is reported to illustrate the usefulness of our approach, in dealing with outliers. Finally, semiparametric proposed models and parametric normal model are compared, graphically with the posterior distribution density of the coefficients. (C) 2009 Elsevier Inc. All rights reserved.

Predicting random effects with an expanded finite population mixed model

Prediction of random effects is an important problem with expanding applications. In the simplest context, the problem corresponds to prediction of the latent value (the mean) of a realized cluster selected via two-stage sampling. Recently, Stanek and Singer [Predicting random effects from finite population clustered samples with response error. J. Amer. Statist. Assoc. 99, 119-130] developed best linear unbiased predictors (BLUP) under a finite population mixed model that outperform BLUPs from mixed models and superpopulation models. Their setup, however, does not allow for unequally sized clusters. To overcome this drawback, we consider an expanded finite population mixed model based on a larger set of random variables that span a higher dimensional space than those typically applied to such problems. We show that BLUPs for linear combinations of the realized cluster means derived under such a model have considerably smaller mean squared error (MSE) than those obtained from mixed models, superpopulation models, and finite population mixed models. We motivate our general approach by an example developed for two-stage cluster sampling and show that it faithfully captures the stochastic aspects of sampling in the problem. We also consider simulation studies to illustrate the increased accuracy of the BLUP obtained under the expanded finite population mixed model. (C) 2007 Elsevier B.V. All rights reserved.

Prevalence of hepatitis B and C serological markers among first-time blood donors in Brazil: A multi-center serosurvey

Little data are available on the seroprevalence of, and risk factors for hepatitis B and C viruses (HBV and HCV) infection in Latin American countries. A multi-center serosurvey was conducted among 3,598 first-time blood donors (65% men) from Sao Paulo, Salvador and Manaus in Brazil. The gender-specific seroprevalences of antibodies against hepatitis B core antigen (anti-HBc) and of the hepatitis B surface antigen (HBsAg) in anti-HBc-positive sera were measured, and risk factors analyzed by gender. The gender-specific seroprevalences of antibodies against HCV (anti-HCV) were measured, but risk factors for HCV were not determined. Anti-HBc and HBsAg seroprevalences were not significantly different in men [101/2,341 (4.31%) and 4/2,229 (0.18%), respectively] and women [65/1,237 (5.25%) and 8/ 1,169 (0.68%), respectively], whereas the seroprevalence of anti-HCV was higher in women (12/1,238 [0.97%] vs. 9/2,353 [0.38%]; odds ratio [OR] = 2.49; 95% confidence interval [Cl]: 1.0-6.0). No significant difference for HBV infection was found across the three study sites or by ethnic group. The seroprevalence of anti-HBc increased with age, but decreased with education level in both genders. Lifetime number of sexual partners was associated with anti-HBc prevalence among men (OR = 1.95; 95% Cl: 1.2-3.1), but not women. The seroprevalence of HBV and HCV was low among Brazilian blood donors, and exposure increased with age in both genders.

A reduction of CETP activity, not an increase, is associated with modestly impaired postprandial lipemia and increased HDL-Cholesterol in adult asymptomatic women

Background: The relationship between CETP and postprandial hyperlipemia is still unclear. We verified the effects of varying activities of plasma CETP on postprandial lipemia and precocious atherosclerosis in asymptomatic adult women. Methods: Twenty-eight women, selected from a healthy population sample (n = 148) were classified according to three CETP levels, all statistically different: CETP deficiency (CETPd <= 4.5%, n = 8), high activity (CETPi >= 23.8, n = 6) and controls (CTL, CETP >= 4.6% and <= 23.7%, n = 14). After a 12 h fast they underwent an oral fat tolerance test (40 g of fat/m(2) of body surface area) for 8 hours. TG, TG-rich-lipoproteins (TRL), cholesterol and TRL-TG measurements (AUC, AUIC, AR, RR and late peaks) and comparisons were performed on all time points. Lipases and phospholipids transfer protein (PLTP) were determined. Correlation between carotid atherosclerosis (c-IMT) and postprandial parameters was determined. CETP TaqIB and I405V and ApoE-epsilon 3/epsilon 2/epsilon 4 polymorphisms were examined. To elucidate the regulation of increased lipemia in CETPd a multiple linear regression analysis was performed. Results: In the CETPi and CTL groups, CETP activity was respectively 9 and 5.3 higher compared to the CETPd group. Concentrations of all HDL fractions and ApoA-I were higher in the CETPd group and clearance was delayed, as demonstrated by modified lipemia parameters (AUC, AUIC, RR, AR and late peaks and meal response patterns). LPL or HL deficiencies were not observed. No genetic determinants of CETP deficiency or of postprandial lipemia were found. Correlations with c-IMT in the CETPd group indicated postprandial pro-atherogenic associations. In CETPd the regression multivariate analysis (model A) showed that CETP was largely and negatively predicted by VLDL-C lipemia (R(2) = 92%) and much less by TG, LDL-C, ApoAI, phospholipids and non-HDL-C. CETP (model B) influenced mainly the increment in ApoB-100 containing lipoproteins (R(2) = 85% negatively) and phospholipids (R(2) = 13%), at the 6(th)h point. Conclusion: The moderate CETP deficiency phenotype included a paradoxically high HDL-C and its sub fractions (as earlier described), positive associations with c-IMT, a postprandial VLDL-C increment predicting negatively CETP activity and CETP activity regulating inversely the increment in ApoB100-containing lipoproteins. We hypothesize that the enrichment of TG content in triglyceride-rich ApoB-containing lipoproteins and in TG rich remnants increases lipoproteins` competition to active lipolysis sites, reducing their catabolism and resulting on postprandial lipemia with atherogenic consequences.

Accumulation of (5 ` S)-8,5 `-cyclo-2 `-deoxyadenosine in organs of Cockayne syndrome complementation group B gene knockout mice

Cockayne syndrome (CS) is a human genetic disorder characterized by sensitivity to UV radiation, neurodegeneration, premature aging among other phenotypes. CS complementation group B (CS-B) gene (csb) encodes the CSB protein (CSB) that is involved in base excision repair of a number of oxidatively induced lesions in genomic DNA in vivo. We hypothesized that CSB may also play a role in cellular repair of the DNA helix-distorting tandem lesion (5`S)-8,5`-cyclo-2`-deoxyadenosine (S-cdA). Among many DNA lesions. S-cdA is unique in that it represents a concomitant damage to both the sugar and base moieties of the same nucleoside. Because of the presence of the C8-C5` covalent bond, S-cdA is repaired by nucleotide excision repair unlike most of other oxidatively induced lesions in DNA, which are subject to base excision repair. To test our hypothesis, we isolated genomic DNA from brain, kidney and liver of wild type and csb knockout (csb(-/-)) mice. Animals were not exposed to any exogenous oxidative stress before the experiment. DNA samples were analysed by liquid chromatography/mass spectrometry with isotope-dilution. Statistically greater background levels of S-cdA were observed in all three organs of csb(-/-) mice than in those of wild type mice. These results suggest the in vivo accumulation of S-cdA in genomic DNA due to lack of its repair in csb(-/-) mice. Thus, this study provides, for the first time, the evidence that CSB plays a role in the repair of the DNA helix-distorting tandem lesion S-cdA. Accumulation of unrepaired S-cdA in vivo may contribute to the pathology associated with CS. Published by Elsevier B.V.

Investigation of Iron (III) desferrioxamine B oxalate ligand exchange by UV-vis spectroscopy: the next step in developing a quantitative analytical method for measuring bioavailable iron in marine ecosystems

To date there are no analytical techniques designed to exclusively measure bioavailable iron in marine environments. The goal of this research is to develop such a technique by isolating the bioavailable iron using the terrestrial siderophore desferrioxamine B (DFB). This project contained many challenging aspects, but the specific goal of this study was to develop a robust analytical technique for quantification of Fe(III)-DFB complexes at nanomolar concentrations. Past work showed that oxalate (Ox) promotes photodissociation of Fe(III)-DFB to Fe(Il), and we are specifically interested in the mechanism of this process. A model was developed using known thermodynamic constants for Fe(III)-DFB and Fe(III) oxalato complexes and adjusting for ionic strength. The model was confirmed by monitoring the UV-VIS absorbance of the system at a variety of oxalate concentrations and pH. The model did not include ternary complexes. Next., the rate of Fe(1I) production during UV irradiation was examined. The results showed that the rate of Fe(II) production was based entirely on the [Fe(Ox)?]3- speciation, and that reoxidation of Fe(II) occurred via reactive oxygen intermediates. This reoxidation could be avoided by either decreasing the oxygen concentration or by adding a Fe(II) stabilizing reagent, such as ferrozine. Further studies need to be done to confirm that these results apply at sub nanomolar concentrations, and the issue of Fe(II) reoxidation at lower Fe concentrations needs to be addressed.

Mapping of Diepoxybutane Damage in the Cytochrome b Domain of Mitochondrial DNA and the β-globin Domain of Nuclear Chicken DNA Using Quantitative PCR

Diepoxybutane (DEB), a known industrial carcinogen, reacts with DNA primarily at the N7 position of deoxyguanosine residues and creates interstrand cross-links at the sequence 5'-GNC. Since N7-N7 cross-links cause DNA to fragment upon heating, quantative polymerase chain reaction (QPCR) is being used in this experiment to measure the amount of DEB damage (lesion frequency) with three different targets-mitochondrial (unpackaged), open chromatin region, and closed chromatin region. Initial measurements of DEB damage within these three targets were not consistent because the template DNA was not the limiting reagent in the PCR. Follow-up PCR trials using a limiting amount of DNA are still in progress although initial experimentation looks promising. Sequencing of these three targets to confirm the primer targets has only been successfully performed for the closed chromatin target and does not match the sequence from NIH used to design that primer pair. Further sequencing trials need to be conducted on all three targets to assure that a mitochondrial, open chromatin, and closed chromatin region are actually being amplified in this experimental series.