EMLasso: logistic lasso with missing data.

In clinical settings, missing data in the covariates occur frequently. For example, some markers are expensive or hard to measure. When this sort of data is used for model selection, the missingness is often resolved through a complete case analysis or a form of single imputation. An alternative sometimes comes in the form of leaving the most damaged covariates out. All these strategies jeopardise the goal of model selection. In earlier work, we have applied the logistic Lasso in combination with multiple imputation to obtain results in such settings, but we only provided heuristic arguments to advocate the method. In this paper, we propose an improved method that builds on firm statistical arguments and that is developed along the lines of the stochastic expectation-maximisation algorithm. We show that our method can be used to handle missing data in both categorical and continuous predictors, as well as in a nonpenalised regression. We demonstrate the method by applying it to data of 273 lung cancer patients. The objective is to select a model for the prediction of acute dysphagia, starting from a large set of potential predictors, including clinical and treatment covariates as well as a set of single-nucleotide polymorphisms.

Biodiversity of freshwater diatom communities during 1000 years of metal mining, land use, and climate change in central Sweden.

We subjected a unique set of high-quality paleoecological data to statistical modeling to examine if the biological richness and evenness of freshwater diatom communities in the Falun area, a historical copper (Cu) mining region in central Sweden, was negatively influenced by 1000 years of metal exposure. Contrary to ecotoxicological predictions, we found no negative relation between biodiversity and the sedimentary concentrations of eight metals. Strikingly, our analysis listed metals (Co, Fe, Cu, Zn, Cd, Pb) or the fractional land cover of cultivated crops, meadow, and herbs indicating land disturbance as potentially promoting biodiversity. However, correlation between metal- and land-cover trends prevented concluding which of these two covariate types positively affected biodiversity. Because historical aqueous metal concentrations--inferred from solid-water partitioning--approached experimental toxicity thresholds for freshwater algae, positive effects of metal mining on biodiversity are unlikely. Instead, the positive relationship between biodiversity and historical land-cover change can be explained by the increasing proportion of opportunistic species when anthropogenic disturbance intensifies. Our analysis illustrates that focusing on the direct toxic effects of metals alone may yield inaccurate environmental assessments on time scales relevant for biodiversity conservation.

The potential for adaptation in a natural Daphnia magna population: broad and narrow-sense heritability of net reproductive rate under Cd stress at two temperatures.

The existence of genetic variability is a key element of the adaptive potential of a natural population to stress. In this study we estimated the additive and non-additive components of the genetic variability of net reproductive rate (R(0)) in a natural Daphnia magna population exposed to Cd stress at two different temperatures. To this end, life-table experiments were conducted with 20 parental and 39 offspring clonal lineages following a 2 × 2 design with Cd concentration (control vs. 3.7 µg Cd/L) and temperature (20 vs. 24 °C) as factors. Offspring lineages were obtained through inter-clonal crossing of the different parental lineages. The population mean, additive and non-additive genetic components of variation in each treatment were estimated by fitting an Animal Model to the observed R(0) values using restricted maximum likelihood estimation. From those estimates broad-sense heritabilities (H(2)), narrow-sense heritabilities (h(2)), total (CV(G)) and additive genetic coefficients of variation (CV(A)) of R(0) were calculated. The exposure to Cd imposed a considerable level of stress to the population, as shown by the fact that the population mean of R(0) exposed to Cd was significantly lower than in the control at the corresponding temperature, i.e. by 23 % at 20 °C and by 88 % at 24 °C. The latter difference indicates that increasing temperature increased the stress level imposed by Cd. The H² and CV(G) were significantly greater than 0 in all treatments, suggesting that there is a considerable degree of genetic determination of R(0) in this population and that clonal selection could rapidly lead to increasing population mean fitness under all investigated conditions. More specifically, the H² was 0.392 at 20 °C+Cd and 0.563 at 24 °C+Cd; the CV(G) was 30.0 % at 20 °C+Cd and was significantly higher (147.6 %) in the 24 °C+Cd treatment. Significant values of h(2) (= 0.23) and CV(A) (= 89.7 %) were only found in the 24 °C+Cd treatment, suggesting that the ability to produce more offspring under this stressful condition may be inherited across sexual generations. In contrast, in the less stressful 20 °C+Cd treatment the h(2) (0.06) and CV(A) (7.0 %) were very low and not significantly higher than zero. Collectively our data indicate that both the asexual and sexual reproduction phases in cyclic parthenogenetic D. magna populations may play a role in the long-term adaptive potential of Daphnia populations to chemical stress (with Cd as the current example) and that environmental variables which influence the stress level of that chemical may influence this adaptive potential (with temperature as current example).

The ARIA guidelines in specialist practice: a nationwide survey.

PROBLEM: In 2001, the ARIA guidelines were published to assist healthcare practitioners in managing allergic rhinitis (AR) according to the best evidence. Very limited information, however, is avail-able on the impact of these guidelines on clinical practice. METHODS: All Belgian Otorhinolaryngologists were invited to complete a questionnaire, covering demographic and professional characteristics, knowledge, use and perception of the ARIA guidelines and 4 clinical case scenarios of AR. RESULTS: Of the 258 (44%) Belgian Otorhinolaryngologists who participated, almost 90% had ever heard about ARIA and 64% had followed a lecture specifically dedicated to the ARIA guidelines. Furthermore, 62% stated to always or mostly follow the ARIA treatment algorithms in the daily management of AR patients. In the clinical case section, adherence to the ARIA guidelines raised with increased self-reported knowledge and use of the ARIA guidelines and among participants that considered the guidelines more userfriendly. Of the respondents, 51% were considered as good com-pliers. Younger age was a significant predictor for good compliance. CONCLUSION: More efforts are required to improve the translation of scientific knowledge into clinical practice and to further identify which factors may influence guideline compliance.

Multiple putative oncogenes at the chromosome 20q amplicon contribute to colorectal adenoma to carcinoma progression.

OBJECTIVE: This study aimed to identify the oncogenes at 20q involved in colorectal adenoma to carcinoma progression by measuring the effect of 20q gain on mRNA expression of genes in this amplicon. METHODS: Segmentation of DNA copy number changes on 20q was performed by array CGH (comparative genomic hybridisation) in 34 non-progressed colorectal adenomas, 41 progressed adenomas (ie, adenomas that present a focus of cancer) and 33 adenocarcinomas. Moreover, a robust analysis of altered expression of genes in these segments was performed by microarray analysis in 37 adenomas and 31 adenocarcinomas. Protein expression was evaluated by immunohistochemistry on tissue microarrays. RESULTS: The genes C20orf24, AURKA, RNPC1, TH1L, ADRM1, C20orf20 and TCFL5, mapping at 20q, were significantly overexpressed in carcinomas compared with adenomas as a consequence of copy number gain of 20q. CONCLUSION: This approach revealed C20orf24, AURKA, RNPC1, TH1L, ADRM1, C20orf20 and TCFL5 genes to be important in chromosomal instability-related adenoma to carcinoma progression. These genes therefore may serve as highly specific biomarkers for colorectal cancer with potential clinical applications.

Clinical diseases in pet black-tailed prairie dogs (Cynomys ludovicianus): a retrospective study in 206 animals.

OBJECTIVES: To describe the health disorders seen in captivity in 206 pet prairie dogs. MATERIALS AND METHODS: Review of medical records of pet prairie dogs that were presented to a veterinary clinic in the period January 2002 to December 2011. RESULTS: The most frequent diseases were respiratory (28.2%), dental (25.2%), integumentary (22.3%) and digestive (19.4%). The most common clinical disorders were rhinitis, elodontoma, hypothermia, gastrointestinal stasis, diarrhoea, obesity and gastrointestinal tympany. CLINICAL SIGNIFICANCE: This listing of clinical diseases will help clinicians focus their disease investigation and increase clinical awareness of prairie dog diseases. There was a high rate of poor or unsuitable husbandry and this was often associated with clinical disease. Of the diseases seen commonly in this study, only elodontoma has been previously reported as being common.

Tests and diagnostic plots for detecting lack-of-fit for circular-linear regression models.

Regression diagnostics and lack-of-fit tests mainly focus on linear-linear regression models. When the design points are distributed on the circumference of a circle, difficulties arise as there is no natural starting point or origin. Most classical lack-of-fit tests require an arbitrarily chosen origin, but different choices may result in different conclusions. We propose a graphical diagnostic tool and a closely related lack-of-fit test, which does not require a natural starting point. The method is based on regional residuals which are defined on arcs of the circle. The graphical method formally locates and visualizes subsets of poorly fitting observations on the circle. A data example from the food technology is used to point out the before-mentioned problems with conventional lack-of-fit tests and to illustrate the strength of the methodology based on regional residuals in detecting and localizing departures from the no-effect hypothesis. A small simulation study shows a good performance of the regional residual test in case of both global and local deviations from the null model. Finally, the ideas are extended to the case of more than one predictor variable.

Spatio-temporal statistical models for river monitoring networks.

When introducing new wastewater treatment plants (WWTP), investors and policy makers often want to know if there indeed is a beneficial effect of the installation of a WWTP on the river water quality. Such an effect can be established in time as well as in space. Since both temporal and spatial components affect the output of a monitoring network, their dependence structure has to be modelled. River water quality data typically come from a river monitoring network for which the spatial dependence structure is unidirectional. Thus the traditional spatio-temporal models are not appropriate, as they cannot take advantage of this directional information. In this paper, a state-space model is presented in which the spatial dependence of the state variable is represented by a directed acyclic graph, and the temporal dependence by a first-order autoregressive process. The state-space model is extended with a linear model for the mean to estimate the effect of the activation of a WWTP on the dissolved oxygen concentration downstream.

Accurate particle size distribution determination by nanoparticle tracking analysis based on 2-D Brownian dynamics simulation.

A physical model is presented to simulate the average step length distribution during nanoparticle tracking analysis experiments as a function of the particle size distribution and the distribution of the number of steps within the tracks. Considering only tracks of at least five steps, numerical simulation could be replaced by a normal distribution approximation. Based on this model, simulation of a step length distribution allows obtaining a much more reliable estimation of the particle size distribution, thereby reducing the artificial broadening of the distribution, as is typically observed by direct conversion of step length to particle size data. As this fitting procedure also allowed including data from particles that were followed for a relatively low number of steps, the measurement time could be reduced for particles that are known to be monodisperse. Whereas the inversion is less sensitive towards the particle size distribution width, still similar values were obtained for both the average diameter and standard deviation of a polystyrene latex sample irrespective of the track length, provided that the latter included at least five steps.

Comparing telomere length of sister chromatids in human lymphocytes using three-dimensional confocal microscopy.

BACKGROUND: The length of the terminal sequences of linear chromosomes changes dynamically during cellular proliferation. A crucial element in the study of telomere-related regulation mechanisms is the ability to measure telomere lengths of individual chromosomes. Individual telomere lengths can be measured using digital imaging fluorescence microscopy-based techniques. We extended this method using confocal microscopy for the acquisition of three-dimensional (3D) images. Consequently, variations in measured signal intensities due to erroneous focusing are avoided. METHODS: We employed our 3D telomere sizing method to compare telomere lengths of sister chromatids within metaphase preparations from human lymphocytes. The samples were treated following a quantitative fluorescence in situ hybridization (Q-FISH) protocol using fluorescein isothiocyanate (FITC)-labeled telomeric peptidic nucleic acid (PNA) probes and propidium iodide (PI) counterstain. RESULTS: We demonstrated that the telomere lengths of two sister chromatids are not necessarily equal in human lymphocytes. Profound statistical analysis demonstrated significant differences in the distribution of the sister chromatid telomere lengths, but we were not able to prove a discrete distribution of telomere sister ratios. These telomere length differences were more apparent in older individuals. CONCLUSION: Whereas the majority of sister telomere pairs have equal lengths, surprisingly, a minority was significantly different in each individual studied. We are convinced that these observations are not linked to the methodology or the protocol applied. We suggest that a biological phenomenon might be involved.

Comparison of computer-controlled administration of propofol with two manually controlled infusion techniques.

Ninety women were studied in order to compare dose requirements and quality of anaesthesia between target-controlled infusion and two manually controlled infusion schemes for propofol administration: group I received target-controlled infusion for induction (4 micrograms.ml-1 target blood concentration, increased by 2 micrograms.ml-1 after 3 min of consciousness not lost), groups II and III received an induction bolus of propofol at infusion rates of 1200 or 600 ml.h-1, respectively, until loss of consciousness. Anaesthesia was maintained with propofol target-controlled infusion in group 1 or by constant rate infusion in the other two groups. Computer simulations were used to calculate blood and effect-site propofol concentrations. Mean induction times (SD) were 78 (65)s in group I versus 51 (10)s and 62 (12)s in groups II and III, respectively (p < 0.05 between groups II and III). Mean induction doses were: 1.31 (0.44), 2.74 (0.56) and 1.77 (0.43) mg.kg-1 and mean maintenance doses were 13.4 (3.55), 9.32 (1.72) and 9.97 (1.53) mg.kg-1 h-1 in groups I, II and III, respectively (p < 0.05 between all groups). There was a lower incidence of apnoea in group I than in groups II and III. There were no significant differences between the groups in other objective parameters of anaesthetic quality studied. Computer simulations showed an "overshoot' in propofol blood and effect-site concentration with manual induction and significantly higher maintenance levels with target-controlled infusion.

Four-dimensional imaging and computer-assisted track analysis of nuclear migration in root hairs of Arabidopsis thaliana.

Nuclear migration is a fundamental mechanism necessary for the proper growth and development of many eukaryotic organisms. In this study root hairs of Arabidopsis thaliana were used as a research model to gain insight into the dynamics of nuclear migration. Root hairs are long tubular outgrowths of epidermal cells and are responsible for the uptake of water and nutrients. During the development of root hairs, the nucleus migrates into the hair after the bulge is formed. The position of the nucleus relative to the tip plays an essential role in the growth process. However, what is happening to the nucleus in full-grown root hairs is still unclear. To study nuclear dynamics in living root hair cells, stably transformed plants with the fusion proteins Histone2B-YFP and NLS-GFP-GUS were used. Four-dimensional confocal laser scanning microscopy made it possible to monitor the exact position of the nucleus in different root hairs. To analyse the sequential positions of the nuclei in the root hairs, a new computer-assisted method was developed. After track analysis a number of parameters could be extracted from the movies, such as the average speed, the amplitude, direction factor and the range of movement in the root hairs. Our results show that nuclei do not reach a final position in full-grown root hairs and this sustained movement seems to be more similar in root hairs lying close to each other. Moreover, with this methodology it could be quantitatively demonstrated that the integrity of actin is necessary for nuclear movement.