Double generalized linear model for tissue culture proportion data: a Bayesian perspective

Joint generalized linear models and double generalized linear models (DGLMs) were designed to model outcomes for which the variability can be explained using factors and/or covariates. When such factors operate, the usual normal regression models, which inherently exhibit constant variance, will under-represent variation in the data and hence may lead to erroneous inferences. For count and proportion data, such noise factors can generate a so-called overdispersion effect, and the use of binomial and Poisson models underestimates the variability and, consequently, incorrectly indicate significant effects. In this manuscript, we propose a DGLM from a Bayesian perspective, focusing on the case of proportion data, where the overdispersion can be modeled using a random effect that depends on some noise factors. The posterior joint density function was sampled using Monte Carlo Markov Chain algorithms, allowing inferences over the model parameters. An application to a data set on apple tissue culture is presented, for which it is shown that the Bayesian approach is quite feasible, even when limited prior information is available, thereby generating valuable insight for the researcher about its experimental results.

Post-transplant recurrent hepatitis C: immunohistochemical detection of hepatitis C virus core antigen and possible pathogenic implications

Introduction: The mechanisms by which severe cholestatic hepatitis develops after liver transplantation are not fully understood. Reports on immunohistochemical distribution of hepatitis C virus (HCV) antigens are still scarce, but recently, HCV immunostaining was suggested for early diagnosis of cholestatic forms of recurrent hepatitis C in liver grafts. After purification, Rb246 pab anticore (aa1-68) yielded specific, granular cytoplasmic staining in hepatocytes. Signal amplification through the Envision-Alkaline Phosphatase System avoided endogenous biotin and peroxidase. Aims/Methods: Rb246 was applied to liver samples of explants of 12 transplant recipients, six with the most severe form of post-transplantation recurrence, severe cholestatic hepatitis (group 1) and six with mild recurrence (group 2). We also assessed immuno-reactivity at two time-points post-transplantation (median 4 and 22 months) in both groups. HCV-core Ag was semiquantified from 0 to 3+ in each time point. Serum HCV-RNA was also measured on the different time points by branched DNA. Results: In the early post-transplant time point, one patient had a mild staining (1+), two patients had a moderate staining (2+) and the other three had no staining in group 1, compared with five patients with no staining (0) and one patient with mild staining (1+) in group 2. Late post-transplant liver samples were available in nine patients, and two out of four samples in group 1 showed a mild staining, compared with no staining patients in five patients in group 2. Strikingly, on the explant samples, HCV immunostaining was strongly positive in group 1, and mildly positive in group 2. Two out of five samples showed 3+ staining, and three samples showed 2+ staining in group 1; two out of five samples showed no staining, two samples showed 1+ staining and one sample showed 2+ staining in group 2. Serum HCV-RNA was significantly higher in group 1, on both time-points post-transplantation. HCV-core Ag was not directly associated with serum HCV-RNA on the different time points. Conclusion: These preliminary results suggest that strong HCV immunostaining in the explant is predictive of more severe disease recurrence.

Matrix metalloproteinases cleave the beta(2)-adrenergic receptor in spontaneously hypertensive rats

Rodrigues SF, Tran ED, Fortes ZB, Schmid-Schonbein GW. Matrix metalloproteinases cleave the beta(2)-adrenergic receptor in spontaneously hypertensive rats. Am J Physiol Heart Circ Physiol 299: H25-H35, 2010. First published April 9, 2010; doi:10.1152/ajpheart.00620.2009.-We recently observed the enhanced serine and matrix metalloproteinase (MMP) activity in the spontaneously hypertensive rat (SHR) compared with its normotensive Wistar-Kyoto (WKY) rat and the cleavage of membrane receptors in the SHR by MMPs. We demonstrate in vivo that MMP-7 and MMP-9 injection leads to a vasoconstrictor response in microvessels of rats that is blocked by a specific MMP inhibitor (GM-6001, 1 mu M). Multiple pathways may be responsible. Since the beta(2)-adrenergic receptor (beta(2)-AR) is susceptible to the action of endogenous MMPs, we hypothesize that MMPs in the plasma of SHRs are able to cleave the extracellular domain of the beta(2)-AR. SHR arterioles respond in an attenuated fashion to beta(2)-AR agonists and antagonists. Aorta and heart muscle of control Wistar rats were exposed for 24 h (37 C) to fresh plasma of male Wistar and WKY rats and SHRs with and without doxycycline (30 mu M) and EDTA (10 mM) to reduce MMP activity. The density of extracellular and intracellular domains of beta(2)-AR was determined by immunohistochemistry. The density of the extracellular domain of beta(2)-AR is reduced in aortic endothelial cells and cardiac microvessels of SHRs compared with that of WKY or Wistar rats. Treatment of the aorta and the heart of control Wistar rats with plasma from SHRs, but not from WKY rats, reduced the number of extracellular domains, but not intracellular domains, of beta(2)-AR in aortic endothelial cells and cardiac microvessels. MMP inhibitors (EDTA and doxycycline) prevented the cleavage of the extracellular domain. Thus MMPs may contribute to the reduced density of the extracellular domain of beta(2)-AR in blood vessels and to the increased arteriolar tone of SHRs compared with normotensive rats.

Inferential Implications of Over-Parametrization: A Case Study in Incomplete Categorical Data

P>In the context of either Bayesian or classical sensitivity analyses of over-parametrized models for incomplete categorical data, it is well known that prior-dependence on posterior inferences of nonidentifiable parameters or that too parsimonious over-parametrized models may lead to erroneous conclusions. Nevertheless, some authors either pay no attention to which parameters are nonidentifiable or do not appropriately account for possible prior-dependence. We review the literature on this topic and consider simple examples to emphasize that in both inferential frameworks, the subjective components can influence results in nontrivial ways, irrespectively of the sample size. Specifically, we show that prior distributions commonly regarded as slightly informative or noninformative may actually be too informative for nonidentifiable parameters, and that the choice of over-parametrized models may drastically impact the results, suggesting that a careful examination of their effects should be considered before drawing conclusions.Resume Que ce soit dans un cadre Bayesien ou classique, il est bien connu que la surparametrisation, dans les modeles pour donnees categorielles incompletes, peut conduire a des conclusions erronees. Cependant, certains auteurs persistent a negliger les problemes lies a la presence de parametres non identifies. Nous passons en revue la litterature dans ce domaine, et considerons quelques exemples surparametres simples dans lesquels les elements subjectifs influencent de facon non negligeable les resultats, independamment de la taille des echantillons. Plus precisement, nous montrons comment des a priori consideres comme peu ou non-informatifs peuvent se reveler extremement informatifs en ce qui concerne les parametres non identifies, et que le recours a des modeles surparametres peut avoir sur les conclusions finales un impact considerable. Ceci suggere un examen tres attentif de l`impact potentiel des a priori.