Accelerating the Pool-Adjacent-Violators Algorithm for Isotonic Distributional Regression.

In the context of estimating stochastically ordered distribution functions, the pool-adjacent-violators algorithm (PAVA) can be modified such that the computation times are reduced substantially. This is achieved by studying the dependence of antitonic weighted least squares fits on the response vector to be approximated.

Wood combustion particles induce adverse effects to normal and diseased airway epithelia.

Residential wood burning is a major source of poorly characterized, deleterious particulate matter, whose composition and toxicity may vary with wood type, burning condition and photochemical age. The causative link between ambient wood particle constituents and observed adverse health effects is currently lacking. Here we investigate the relationship between chemical properties of primary and atmospherically aged wood combustion particles and acute toxicity in human airway epithelial cells. Emissions from a log wood burner were diluted and injected into a smog chamber for photochemical aging. After concentration-enrichment and removal of oxidizing gases, directly emitted and atmospherically aged particles were deposited on cell cultures at the air-liquid interface for 2 hours in an aerosol deposition chamber mimicking physiological conditions in lungs. Cell models were fully differentiated normal and diseased (cystic fibrosis and asthma) human bronchial epithelia (HBE) and the bronchial epithelial cell line BEAS-2B. Cell responses were assessed at 24 hours after aerosol exposure. Atmospherically relevant doses of wood combustion particles significantly increased cell death in all but the asthma cell model. Expression of oxidative stress markers increased in HBE from all donors. Increased cell death and inflammatory responses could not be assigned to a single chemical fraction of the particles. Exposure to primary and aged wood combustion particles caused adverse effects to airway epithelia, apparently induced by several interacting components.

A law of the iterated logarithm for Grenander's estimator.

In this note we prove the following law of the iterated logarithm for the Grenander estimator of a monotone decreasing density: If f(t0) > 0, f'(t0) < 0, and f' is continuous in a neighborhood of t0, then [Formula: see text]almost surely where [Formula: see text]here [Formula: see text] is the two-sided Strassen limit set on [Formula: see text]. The proof relies on laws of the iterated logarithm for local empirical processes, Groeneboom's switching relation, and properties of Strassen's limit set analogous to distributional properties of Brownian motion.

Toxicity of aged gasoline exhaust particles to normal and diseased airway epithelia.

Particulate matter (PM) pollution is a leading cause of premature death, particularly in those with pre-existing lung disease. A causative link between particle properties and adverse health effects remains unestablished mainly due to complex and variable physico-chemical PM parameters. Controlled laboratory experiments are required. Generating atmospherically realistic aerosols and performing cell-exposure studies at relevant particle-doses are challenging. Here we examine gasoline-exhaust particle toxicity from a Euro-5 passenger car in a uniquely realistic exposure scenario, combining a smog chamber simulating atmospheric ageing, an aerosol enrichment system varying particle number concentration independent of particle chemistry, and an aerosol deposition chamber physiologically delivering particles on air-liquid interface (ALI) cultures reproducing normal and susceptible health status. Gasoline-exhaust is an important PM source with largely unknown health effects. We investigated acute responses of fully-differentiated normal, distressed (antibiotics-treated) normal, and cystic fibrosis human bronchial epithelia (HBE), and a proliferating, single-cell type bronchial epithelial cell-line (BEAS-2B). We show that a single, short-term exposure to realistic doses of atmospherically-aged gasoline-exhaust particles impairs epithelial key-defence mechanisms, rendering it more vulnerable to subsequent hazards. We establish dose-response curves at realistic particle-concentration levels. Significant differences between cell models suggest the use of fully-differentiated HBE is most appropriate in future toxicity studies.

A Drosophila XPD model links cell cycle coordination with neuro-development and suggests links to cancer.

XPD functions in transcription, DNA repair and in cell cycle control. Mutations in human XPD (also known as ERCC2) mainly cause three clinical phenotypes: xeroderma pigmentosum (XP), Cockayne syndrome (XP/CS) and trichothiodystrophy (TTD), and only XP patients have a high predisposition to developing cancer. Hence, we developed a fly model to obtain novel insights into the defects caused by individual hypomorphic alleles identified in human XP-D patients. This model revealed that the mutations that displayed the greatest in vivo UV sensitivity in Drosophila did not correlate with those that led to tumor formation in humans. Immunoprecipitations followed by targeted quantitative MS/MS analysis showed how different xpd mutations affected the formation or stability of different transcription factor IIH (TFIIH) subcomplexes. The XP mutants most clearly linked to high cancer risk, Xpd R683W and R601L, showed a reduced interaction with the core TFIIH and also an abnormal interaction with the Cdk-activating kinase (CAK) complex. Interestingly, these two XP alleles additionally displayed high levels of chromatin loss and free centrosomes during the rapid nuclear division phase of the Drosophila embryo. Finally, the xpd mutations showing defects in the coordination of cell cycle timing during the Drosophila embryonic divisions correlated with those human mutations that cause the neurodevelopmental abnormalities and developmental growth defects observed in XP/CS and TTD patients.

Probability of anti-D development in D- patients receiving D+ RBCs.

BACKGROUND: In some situations, the administration of D+ RBCs to D- patients is necessary. The probability of a subsequent anti-D formation is assumed to be around 80 percent, a figure based primarily on studies in healthy volunteers. It was hypothesized that patients requiring blood transfusion have a much lower probability of developing antibodies. STUDY DESIGN AND METHODS: A retrospective analysis was performed whereby 78 D- patients were evaluated for the development of RBC antibodies after administration of D+ RBCs. For the analysis of the cross-sectional observations, parametric models were used for interval-censored data. RESULTS: Anti-D was detected in 16 of 78 patients. Considering the individual patient's inspection times, the calculated probability of developing antibody following D+ RBC supply was shown to be below 41.7 percent (upper 95% confidence bound) and estimated as 30.4 percent. The data hinted toward an inverse correlation between the number of transfused units and the probability of antibody formation. Interestingly, 6 of these 16 patients developed additional IgG autoantibody. In 3 of those cases, evidence for prolonged hemolysis was found. CONCLUSION: The actual frequency of antibody formation in our patients is much lower than assumed. On the other hand, prolonged hemolysis probably induced by additional autoreactive antibodies might occur. This possible complication has not yet been addressed. Further studies might reveal whether a less restricted transfusion policy with respect to D matching is justified in selected patients.

Detection of cortical transition regions utilizing statistical analyses of excess masses.

A new statistical approach for observer-assisted detection of transition regions of adjacent cytoarchitectonic areas within the human cerebral cortex was developed. This method analyzes the structural information of cytoarchitectural profiles (e.g., the modality of a gray level intensity distribution) based on observed excess mass differences verified by a suitable statistical test. Profiles were generated by scanning the cerebral cortex over respective regions of interest that were oriented to trajectories running parallel to the orientation of cell columns. For each single profile, determination of excess masses provided evidence for a certain number of peaks in the cell density, thereby avoiding fluctuation due solely to sampling anomalies. Comparing such excess mass measurements by means of multiple local rank tests over a wide range of profiles allowed for the detection of cytoarchitectural inhomogeneities at respective given confidence levels. Special parameters (e.g., level of significance, width of targeted region, number of peaks) then could be adapted to specific pattern recognition problems in lamination analyses. Such analyses of excess masses provided a general tool for observer-assisted evaluation of profile arrays. This observer-assisted statistical method was applied to five different cortical examples. It detected the same transition regions that had been determined earlier through direct examination of samples, despite cortical convexities, concavities, and some minor staining inhomogeneities.