Posterior Simulation in the Generalized Linear Model with Semiparmetric Random Effects

Generalized linear mixed models with semiparametric random effects are useful in a wide variety of Bayesian applications. When the random effects arise from a mixture of Dirichlet process (MDP) model, normal base measures and Gibbs sampling procedures based on the Pólya urn scheme are often used to simulate posterior draws. These algorithms are applicable in the conjugate case when (for a normal base measure) the likelihood is normal. In the non-conjugate case, the algorithms proposed by MacEachern and Müller (1998) and Neal (2000) are often applied to generate posterior samples. Some common problems associated with simulation algorithms for non-conjugate MDP models include convergence and mixing difficulties. This paper proposes an algorithm based on the Pólya urn scheme that extends the Gibbs sampling algorithms to non-conjugate models with normal base measures and exponential family likelihoods. The algorithm proceeds by making Laplace approximations to the likelihood function, thereby reducing the procedure to that of conjugate normal MDP models. To ensure the validity of the stationary distribution in the non-conjugate case, the proposals are accepted or rejected by a Metropolis-Hastings step. In the special case where the data are normally distributed, the algorithm is identical to the Gibbs sampler.

Studying Effects of Primary Care Physicians and Patients on the Trade-Off Between Charges for Primary Care and Specialty Care Using a Hierarchical Multivariate Two-Part Model

Objective. To examine effects of primary care physicians (PCPs) and patients on the association between charges for primary care and specialty care in a point-of-service (POS) health plan. Data Source. Claims from 1996 for 3,308 adult male POS plan members, each of whom was assigned to one of the 50 family practitioner-PCPs with the largest POS plan member-loads. Study Design. A hierarchical multivariate two-part model was fitted using a Gibbs sampler to estimate PCPs' effects on patients' annual charges for two types of services, primary care and specialty care, the associations among PCPs' effects, and within-patient associations between charges for the two services. Adjusted Clinical Groups (ACGs) were used to adjust for case-mix. Principal Findings. PCPs with higher case-mix adjusted rates of specialist use were less likely to see their patients at least once during the year (estimated correlation: –.40; 95% CI: –.71, –.008) and provided fewer services to patients that they saw (estimated correlation: –.53; 95% CI: –.77, –.21). Ten of 11 PCPs whose case-mix adjusted effects on primary care charges were significantly less than or greater than zero (p < .05) had estimated, case-mix adjusted effects on specialty care charges that were of opposite sign (but not significantly different than zero). After adjustment for ACG and PCP effects, the within-patient, estimated odds ratio for any use of primary care given any use of specialty care was .57 (95% CI: .45, .73). Conclusions. PCPs and patients contributed independently to a trade-off between utilization of primary care and specialty care. The trade-off appeared to partially offset significant differences in the amount of care provided by PCPs. These findings were possible because we employed a hierarchical multivariate model rather than separate univariate models.

INTERACTING WITH LOCAL AND REMOTE DATA RESPOSITORIES USING THE stashR PACKAGE

The stashR package (a Set of Tools for Administering SHared Repositories) for R implements a simple key-value style database where character string keys are associated with data values. The key-value databases can be either stored locally on the user's computer or accessed remotely via the Internet. Methods specific to the stashR package allow users to share data repositories or access previously created remote data repositories. In particular, methods are available for the S4 classes localDB and remoteDB to insert, retrieve, or delete data from the database as well as to synchronize local copies of the data to the remote version of the database. Users efficiently access information from a remote database by retrieving only the data files indexed by user-specified keys and caching this data in a local copy of the remote database. The local and remote counterparts of the stashR package offer the potential to enhance reproducible research by allowing users of Sweave to cache their R computations for a research paper in a localDB database. This database can then be stored on the Internet as a remoteDB database. When readers of the research paper wish to reproduce the computations involved in creating a specific figure or calculating a specific numeric value, they can access the remoteDB database and obtain the R objects involved in the computation.

Analysis of gene order data supports vertical inheritance of the leukotoxin operon and genome rearrangements in the 5' flanking region in genus Mannheimia

BACKGROUND: The Mannheimia subclades belong to the same bacterial genus, but have taken divergent paths toward their distinct lifestyles. For example, M. haemolytica + M. glucosida are potential pathogens of the respiratory tract in the mammalian suborder Ruminantia, whereas M. ruminalis, the supposed sister group, lives as a commensal in the ovine rumen. We have tested the hypothesis that vertical inheritance of the leukotoxin (lktCABD) operon has occurred from the last common ancestor of genus Mannheimia to any ancestor of the diverging subclades by exploring gene order data. RESULTS: We examined the gene order in the 5' flanking region of the leukotoxin operon and found that the 5' flanking gene strings, hslVU-lapB-artJ-lktC and xylAB-lktC, are peculiar to M. haemolytica + M. glucosida and M. granulomatis, respectively, whereas the gene string hslVU-lapB-lktC is present in M. ruminalis, the supposed sister group of M. haemolytica + M. glucosida, and in the most ancient subclade M. varigena. In M. granulomatis, we found remnants of the gene string hslVU-lapB-lktC in the xylB-lktC intergenic region. CONCLUSION: These observations indicate that the gene string hslVU-lapB-lktC is more ancient than the hslVU-lapB-artJ-lktC and xylAB-lktC gene strings. The presence of (remnants of) the ancient gene string hslVU-lapB-lktC among any subclades within genus Mannheimia supports that it has been vertically inherited from the last common ancestor of genus Mannheimia to any ancestor of the diverging subclades, thus reaffirming the hypothesis of vertical inheritance of the leukotoxin operon. The presence of individual 5' flanking regions in M. haemolytica + M. glucosida and M. granulomatis reflects later genome rearrangements within each subclade. The evolution of the novel 5' flanking region in M. haemolytica + M. glucosida resulted in transcriptional coupling between the divergently arranged artJ and lkt promoters. We propose that the chimeric promoter have led to high level expression of the leukotoxin operon which could explain the increased potential of certain M. haemolytica + M. glucosida strains to cause a particular type of infection.

Cladophora, mass transport, and the nearshore phosphorus shunt

Excessive Cladophora growth in the Great Lakes has led to beach fouling and the temporary closure of nuclear power plants and has been associated with avian botulism and the persistence of human pathogens. As the growth-limiting nutrient for Cladophora, phosphorus is the appropriate target for management efforts. Dreissenids (zebra and quagga mussels) have the ability to capture particulate phase phosphorus (otherwise unavailable to Cladophora) and release it in a soluble, available form. The significance of this potential nutrient source is, in part, influenced by the interplay between phosphorus flux from the mussel bed and turbulent mixing in establishing the phosphorus levels to which Cladophora is exposed. It is hypothesized that under quiescent conditions phosphorus will accumulate near the sediment-water interface, setting up vertical phosphorus gradients and favorable conditions for resource delivery to Cladophora. These gradients would be eliminated under conditions of wind mixing, reducing the significance of the dreissenid-mediated nutrient contribution. Soluble reactive phosphorus (SRP) levels were monitored over dreissenid beds (densities on the order of 350•m-2 and 3000∙m-2) at a site 8 m deep in Lake Michigan. Monitoring was based on the deployment of Modified Hesslein Samplers which collected samples for SRP analysis over a distance of 34 cm above the bottom in 2.5 cm intervals. Deployment intervals were established to capture a wind regime (calm, windy) that persisted for an interval consistent with the sampler equilibration time of 7 hours. Results indicate that increased mussel density leads to an increased concentration boundary layer; increased wind speed leads to entrainment of the concentration boundary layer; and increased duration of quiescent periods leads to an increased concentration boundary layer. This concentration boundary layer is of ecological significance and forms in the region inhabited by Cladophora

The Acropolitan - v. 5, no. 2

In this issue...Professor L. J. Hartzell, Washoe Sampler, Anaconda, Montana, Co-ed Club, Manganese Plant, Professor Gilbert, Ore Diggers, Mines Football, Columbia Gardens

Electroelasticity of charged black branes

We present the first-order corrected dynamics of fluid branes carrying higher-form charge by obtaining the general form of their equations of motion to pole-dipole order in the absence of external forces. Assuming linear response theory, we characterize the corresponding effective theory of stationary bent charged (an)isotropic fluid branes in terms of two sets of response coefficients, the Young modulus and the piezoelectric moduli. We subsequently find large classes of examples in gravity of this effective theory, by constructing stationary strained charged black brane solutions to first order in a derivative expansion. Using solution generating techniques and bent neutral black branes as a seed solution, we obtain a class of charged black brane geometries carrying smeared Maxwell charge in Einstein-Maxwell-dilaton gravity. In the specific case of ten-dimensional space-time we furthermore use T-duality to generate bent black branes with higher-form charge, including smeared D-branes of type II string theory. By subsequently measuring the bending moment and the electric dipole moment which these geometries acquire due to the strain, we uncover that their form is captured by classical electroelasticity theory. In particular, we find that the Young modulus and the piezoelectric moduli of our strained charged black brane solutions are parameterized by a total of 4 response coefficients, both for the isotropic as well as anisotropic cases.

Heterotic T-folds with a small number of neutral moduli

We discuss non-geometric supersymmetric heterotic string models in D=4, in the framework of the free fermionic construction. We perform a systematic scan of models with four a priori left-right asymmetric Z2 projections and shifts. We analyze some 220 models, identifying 18 inequivalent classes and addressing variants generated by discrete torsions. They do not contain geometrical or trivial neutral moduli, apart from the dilaton. However, we show the existence of flat directions in the form of exactly marginal deformations and identify patterns of symmetry breaking where product gauge groups, realized at level one, are broken to their diagonal at higher level. We also describe an “inverse Gepner map” from Heterotic to Type II models that could be used, in certain non geometric settings, to define “effective” topological invariants.

Dirichlet-Neumann inverse spectral problem for a star graph of Stieltjes strings

We solve two inverse spectral problems for star graphs of Stieltjes strings with Dirichlet and Neumann boundary conditions, respectively, at a selected vertex called root. The root is either the central vertex or, in the more challenging problem, a pendant vertex of the star graph. At all other pendant vertices Dirichlet conditions are imposed; at the central vertex, at which a mass may be placed, continuity and Kirchhoff conditions are assumed. We derive conditions on two sets of real numbers to be the spectra of the above Dirichlet and Neumann problems. Our solution for the inverse problems is constructive: we establish algorithms to recover the mass distribution on the star graph (i.e. the point masses and lengths of subintervals between them) from these two spectra and from the lengths of the separate strings. If the root is a pendant vertex, the two spectra uniquely determine the parameters on the main string (i.e. the string incident to the root) if the length of the main string is known. The mass distribution on the other edges need not be unique; the reason for this is the non-uniqueness caused by the non-strict interlacing of the given data in the case when the root is the central vertex. Finally, we relate of our results to tree-patterned matrix inverse problems.

Gene expression in sea urchin development: Study of {\it LpS1\/} gene regulation and cloning and characterization of the {\it SpKrox}-{\it 1\/} gene

Cell differentiation are associated with activation of cell lineage-specific genes. The $LpS{\it 1}\beta$ gene of Lytechinus pictus is activated at the late cleavage stage. $LpS{\it 1}\beta$ transcripts accumulate exclusively in aboral ectoderm lineages. Previous studies demonstrated two G-string DNA-elements, proximal and distal G-strings, which bind to an ectoderm-enriched nuclear factor. In order to define the cis-elements which control positive expression of the $LpS{\it 1}\beta$ gene, the regulatory region from $-$108 to +17 bp of the $LpS{\it 1}\beta$ gene promoter was characterized. The ectoderm G-string factor binds to a G/C-rich region larger than the G-string itself and the binding of the G-string factor requires sequences immediately downstream from the G-string. These downstream sequences are essential for full promoter activity. In addition, only 108 bp of $LpS{\it 1}\beta\ 5\sp\prime$ flanking DNA drives $LpS{\it 1}\beta$ gene expression in aboral ectoderm/mesenchyme cells. Therefore, for positive control of $LpS{\it 1}\beta$ gene expression, two regions of 5$\sp\prime$ flanking DNA are required: region I from base pairs $-$762 to $-$511, and region II, which includes the G/C-rich element, from base pairs $-$108 to $-$61. A mesenchyme cell repressor element is located within region I.^ DNA-binding proteins play key roles in determination of cell differentiation. The zinc finger domain is a DNA-binding domain present in many transcription factors. Based on homologies in zinc fingers, a zinc finger-encoding gene, SpKrox-1, was cloned from S. purpuratus. The putative SpKrox-1 protein has all structural characteristics of a transcription factor: four zinc fingers for DNA binding; acidic domain for transactivation; basic domain for nuclear targeting; and leucine zipper for dimerization. SpKrox-1 RNA transcripts showed a transient expression pattern which correlates largely with early embryonic development. The spatial expression of SpKrox-1 mRNA was distributed throughout the gastrula and larva ectodermal wall. However, SpKrox-1 was not expressed in pigment cells. The SpKrox-1 gene is thus a marker of a subset of SMCs or ectoderm cells. The structural features, and the transient temporal and restricted spatial expression patterns suggest that SpKrox-1 plays a role in a specific developmental event. ^

Sequential updating via dynamic hierarchical models: A Bayesian approach to longitudinal data analysis

Most statistical analysis, theory and practice, is concerned with static models; models with a proposed set of parameters whose values are fixed across observational units. Static models implicitly assume that the quantified relationships remain the same across the design space of the data. While this is reasonable under many circumstances this can be a dangerous assumption when dealing with sequentially ordered data. The mere passage of time always brings fresh considerations and the interrelationships among parameters, or subsets of parameters, may need to be continually revised. ^ When data are gathered sequentially dynamic interim monitoring may be useful as new subject-specific parameters are introduced with each new observational unit. Sequential imputation via dynamic hierarchical models is an efficient strategy for handling missing data and analyzing longitudinal studies. Dynamic conditional independence models offers a flexible framework that exploits the Bayesian updating scheme for capturing the evolution of both the population and individual effects over time. While static models often describe aggregate information well they often do not reflect conflicts in the information at the individual level. Dynamic models prove advantageous over static models in capturing both individual and aggregate trends. Computations for such models can be carried out via the Gibbs sampler. An application using a small sample repeated measures normally distributed growth curve data is presented. ^

Influence of an electromagnetic field on high dilutions measured by ultraviolet light spectroscopy

Background: High dilutions of various starting materials, e.g. copper sulfate, Hypericum perforatum and sulfur, showed significant differences from controls and amongst different dilution levels in ultraviolet light (UV) transmission [1,2]. Exposure of high dilutions to external physical factors such as UV light or elevated temperature (37°C) also yielded significantly different UV transmissions compared to unexposed dilutions [2,3]. In a study with highland frogs it was shown that animals incubated with thyroxine 30c but not with thyroxine 30c exposed to electromagnetic fields (EMFs) of a microwave oven or mobile phone metamorphosed more slowly than control animals [4]. Aims: The aim was to test whether the EMF of a mobile phone influences the UV absorbance of dilutions of quartz and Atropa belladonna (AB). Methodology: Commercially available dilutions of 6x, 12x, 15x, 30x in H2O and 19% ethanol of quartz (SiO2) and of 4x, 6x, 12x, 15x, 30x in H2O and 19% ethanol of AB were used in the experiments (Weleda AG, Arlesheim, Switzerland). Four samples of each dilution were exposed to the EMF of a mobile phone (Philips, Savvy Dual Band) at 900 MHz with an output of 2 W for 3 h, while control samples (4 of each dilution) were kept in a separate room. Absorbance of the samples in the UV range (from 190 to 340 nm) was measured in a randomized order with a Shimadzu UV-1800 spectrophotometer equipped with an auto sampler. In total 5 separate measurement days will be carried out for quartz and for AB dilutions. The average absorbance from 200 to 340 nm and from 200 to 240 nm was compared among dilution levels using a Kruskal-Wallis test and between exposed and unexposed samples using a Mann-Whitney-U test. Results: Preliminary results after 2 measurement days indicated that for quartz the absorbance of the various dilution levels was different from each other (except 12x and 15x), and that samples exposed to an EMF did not show a difference in UV absorbance from unexposed samples. Preliminary results after one measurement day indicated that for AB the absorbance of the various dilution levels was different from each other. The samples exposed to an EMF did not show a difference in UV absorbance from unexposed samples (except 4x in the range from 200 – 240 nm). Conclusions: These results suggest that exposure of high dilutions of quartz and AB to a mobile phone EMF as used here does not alter UV absorbance of these dilutions. The final results will show whether this holds true.

Comparison of globules prepared from high dilutions of various starting materials by ultraviolet light spectroscopy

Background: High and ultra-high dilutions of various starting materials, e.g. copper sulfate, Hypericum perforatum and sulfur, showed significant differences in ultraviolet light (UV) transmission from controls and amongst different dilution levels [1,2]. Verum and placebo globules of Aconitum napellus 30c or calcium carbonate/quercus e cortice 6x from the same packs as used in previous clinical trials and dissolved in water could be distinguished by UV spectroscopy [3]. However, it was unclear whether the differences in UV absorbance originated from specific characteristics of the starting materials, from differences in the production of verum and placebo globules, and/or other unknown interference factors. Aims: The aim of this study was to investigate whether globules produced with high and ultra-high dilutions (6x, 12x, 30c, 200c, 200CF (centesimal discontinuous fluxion), 10,000CF) of various starting materials (Aconitum napellus, Atropa belladonna, phosphorus, sulfur, Apis mellifica, quartz) could be distinguished by UV spectroscopy. Methodology: The globules were specially produced for this study by Spagyros AG (Gümligen, Switzerland) and differed only in the starting materials of the dilutions (but not in the batch of globules or ethanol used). Globules were dissolved in water at 10 mg/ml, in quadruplicates, approximately 22 h prior to the measurements. Absorbance of the samples in the UV range (from 190 to 340 nm) was measured in a randomized order with a Shimadzu double beam UV-1800 spectrophotometer equipped with an auto sampler. Samples of each starting material were prepared and measured on 5 independent days. The daily variations of the spectrophotometer as well as the drift during the measurements were corrected for. The average absorbance from 200 to 340 nm was compared among various starting materials within equal dilution levels using a Kruskal-Wallis test. Results: Statistically significant differences were found among 30c (Figure 1), 200c and 200CF dilutions of the various starting materials. No differences were found among 6x, 12x and 10,000CF dilutions. Conclusions: Globules prepared from high dilutions of various starting materials may show significantly different UV absorbance when dissolved in water. References [1] Wolf U, Wolf M, Heusser P, Thurneysen A, Baumgartner S. Homeopathic preparations of quartz, sulfur and copper sulfate assessed by UV-spectroscopy. Evid Based Complement Alternat Med. 2011;2011:692798. [2] Klein SD, Sandig A, Baumgartner S, Wolf U. Differences in median ultraviolet light transmissions of serial homeopathic dilutions of copper sulfate, Hypericum perforatum, and sulfur. Evid Based Complement Alternat Med. 2013;2013:370609. [3] Klein SD, Wolf U. Investigating homeopathic verum and placebo globules with ultraviolet spectroscopy. Forsch Komplementmed. 2013, accepted.

Possible Demographic Consequences of Intraspecific Shelter Competition Among American Lobsters

Pioneering work by J. Stan Cobb described how habitat architecture and body size scaling affect shelter-related behavior of American lobsters. Subsequent research suggested that shelter availability and competition could set local carrying capacity and demographics for this species. To determine how shelter spacing affects population density, the intensity of intraspecific competition and the distribution of body size for this species, I deployed sets of 10 identically sized artificial shelters spaced at distances of 2.5, 0.5, 1.0, 1.5 and 2.0 meters on otherwise featureless substrate at 10 m depth in mid-coast Maine, U.S.A. Five sets had two parallel strings of five opposing shelters and an additional linear string set 2 to apart without opposing shelters was the most widely separated treatment. Shelters spaced I m apart and closer had higher lobster population densities, more intraspecific competition and higher proportions of empty shelters. Surprisingly, lobsters there were also significantly smaller, declining from 62.7 mm to 50.9 on the carapace (CL) for 2 to linear to 0.25 m spaced shelters, respectively. Nearly all 932 lobsters measured in this study were juvenile (< 90 mm CL) and preharvestable (< 83 mm CL) sized, so mate selection and fishing effects were unlikely. At the scale of the experiment, larger lobsters leave or avoid areas of high lobster population density and intense competition for areas of low population density and relaxed competition (called "demographic diffusion"). Scuba surveys in coastal zones found lobster population densities scale with shelter densities and were highest in boulder habitat where, like the experiment, more than half the shelters were vacant. Fisheries independent scuba and trawl surveys in Maine's shallow coastal zone repeatedly recorded declines of preharvestable, lobsters larger than 60 turn CL in size and increases of those sizes offshore and in deep water. It is possible that this demographic diffusion is driven by behaviors associated with intraspecific shelter competition.

Search for contact interactions and large extra dimensions in dilepton events from pp collisions at sqrts=7 TeV with the ATLAS detector

A search for nonresonant new phenomena, originating from either contact interactions or large extra spatial dimensions, has been carried out using events with two isolated electrons or muons. These events, produced at the LHC in proton-proton collisions at root s = 7 TeV, were recorded by the ATLAS detector. The data sample, collected throughout 2011, corresponds to an integrated luminosity of 4.9 and 5.0 fb(-1) in the e(+)e(-) and mu(+)mu(-) channels, respectively. No significant deviations from the Standard Model expectation are observed. Using a Bayesian approach, 95% confidence level lower limits ranging from 9.0 to 13.9 TeV are placed on the energy scale of llqq contact interactions in the left-left isoscalar model. Lower limits ranging from 2.4 to 3.9 TeV are also set on the string scale in large extra dimension models. After combining these limits with results from a similar search in the diphoton channel, slightly more stringent limits are obtained.