Inference for the Break Point in Segmented Regression with Application to Longitudinal Data

This paper considers inference for the break point in the segmented regression or piece‐wise regression model. Standard likelihood theory does not apply because the break point is absent under the null hypothesis. We use results by Davies for this type of non‐standard set‐up [Biometrika 64 (1977), 247–254 and 74 (1987), 33–43] to obtain a test for the null hypothesis of no break point. A confidence interval can be constructed provided replicate data are available. The methods are exemplified using two longitudinal datasets, the one from ecology, the other from pharmacology.     

Variance Component Estimation for Mixed Model Analysis of cDNA Microarray Data

Microarrays provide a valuable tool for the quantification of gene expression. Usually, however, there is a limited number of replicates leading to unsatisfying variance estimates in a gene‐wise mixed model analysis. As thousands of genes are available, it is desirable to combine information across genes. When more than two tissue types or treatments are to be compared it might be advisable to consider the array effect as random. Then information between arrays may be recovered, which can increase accuracy in estimation. We propose a method of variance component estimation across genes for a linear mixed model with two random effects. The method may be extended to models with more than two random effects. We assume that the variance components follow a log‐normal distribution. Assuming that the sums of squares from the gene‐wise analysis, given the true variance components, follow a scaled χ²‐distribution, we adopt an empirical Bayes approach. The variance components are estimated by the expectation of their posterior distribution. The new method is evaluated in a simulation study. Differentially expressed genes are more likely to be detected by tests based on these variance estimates than by tests based on gene‐wise variance estimates. This effect is most visible in studies with small array numbers. Analyzing a real data set on maize endosperm the method is shown to work well. (© 2008 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Interactions of Bax and tBid with Lipid Monolayers

The release of cytochrome c from mitochondria to the cytosol is a crucial step of apoptosis that involves interactions of Bax and tBid proteins with the mitochondrial membrane. We investigated Bax and tBid interactions with (i) phosphatidylcholine (PC) monolayer as the main component of the outer leaflet of the outer membrane, (ii) with phosphatidylethanolamine (PE) and phosphatidylserine (PS) that are present in the inner leaflet and (iii) with a mixed PC/PE/Cardiolipin (CL) monolayer of the contact sites between the outer and inner membranes. These interactions were studied by measuring the increase of the lipidic monolayer surface pressure induced by the proteins. Our measurements suggest that tBid interacts strongly with the POPC/DOPE/CL, whereas Bax interaction with this monolayer is about 12 times weaker. Both tBid and Bax interact moderately half as strongly with negatively charged DOPS and non-lamellar DOPE monolayers. TBid also slightly interacts with DOPC. Our results suggest that tBid but not Bax interacts with the PC-containing outer membrane. Subsequent insertion of these proteins may occur at the PC/PE/CL sites of contact between the outer and inner membranes. It was also shown that Bax and tBid being mixed in solution inhibit their insertion into POPC/DOPE/CL monolayer. The known 3-D structures of Bax and Bid allowed us to propose a structural interpretation of these experimental results.     

A simple model of plankton population dynamics coupled with a LES of the surface mixed layer

The concentration of phytoplankton in the sea is affected by biological processes, such as growth/mortality rates, predatory zooplankton concentrations and nutrient levels. Phytoplankton concentrations are also influenced by physical processes, in particular the mixing properties of the local fluid environment. On planktonic scales (approximately 10-1000 microm) one can assume the local turbulent flow is isotropic, with no distinction between horizontal and vertical mixing. However, agglomerations of phytoplankton into patches are observed on larger scales of up to hundreds of metres, whose formation will be influenced by the anisotropic advection/mixing properties and large-eddy structures prevalent in the surface mixed layer. This paper presents the results of the coupling of a large-eddy simulation (LES) model of the mixed layer with an advection-diffusion system of coupled equations for nitrate-phytoplankton-zooplankton (NPZ) concentration, incorporating sub-grid parameterizations of the biological processes. Typically these include phytoplankton growth due to light levels and ambient nitrate concentration, offset by grazing losses due to the presence of zooplankton. The primary goal of this work is to investigate how the characteristics of the mixed layer turbulence influence the observed distribution of phytoplankton. One novel feature is the incorporation of a 'vortex-force' term in the LES code in order to generate Langmuir circulations. It has been speculated that the enhanced mixing rates associated with 'Langmuir turbulence' play a significant role in regulating planktonic activity. Results derived from the coupled LES-NPZ model, run with and without the presence of Langmuir circulations, are presented in order to investigate these ideas.     

Arum- and Paris-type arbuscular mycorrhizas in a mixed pine forest on sand dune soil in Niigata Prefecture, central Honshu, Japan

Arbuscular mycorrhizas (AM) are the most widespread mycorrhiza in nature and form two morphologies, Arum- and Paris-type. The determining factors defining the two different morphologies are not well understood. In this study, the distribution of Arum- and Paris-type AM was determined in a mixed pine forest. A total of 35 plant species belonging to 20 families and 32 genera were identified and examined for AM colonization and morphological types. AM morphological types in 14 families were confirmed as follows: Arum-type in Rosaceae, Oleaceae, Lauraceae, Vitaceae and Compositae, Paris-type in Aquifoliaceae, Ulmaceae, Araliaceae, Theaceae, Magnoliaceae, Rubiaceae and Dioscoraceae, and both and/or intermediate types in Caprifoliaceae and Gramineae. Plant families whose AM morphological status was previously unknown were clarified as follows: Polygonaceae and Commelinaceae showed Arum-type morphology; Celastraceae, Menispermaceae and Elaeagnaceae had typical Paris-type morphology. The proportion of Arum-type to Paris-type species decreased in the following order: annuals > perennials > deciduous species > evergreen species, and pioneer group > early successional group > late successional group. Evergreen plants had a higher tendency to form Paris-type AM than annuals, perennials and deciduous plants. The results indicate that environmental changes, such as shade during plant succession, control the distribution of plant growth forms in mixed pine forest and may also play a part in the distribution of Arum- and Paris-type morphology. The identity of the plant seems to strongly influence AM morphology, though control by the fungal genome cannot be ruled out.     

Conditional likelihood score functions for mixed models in linkage analysis

In this paper, we develop a general strategy for linkage analysis, applicable for arbitrary pedigree structures and genetic models with one major gene, polygenes and shared environmental effects. Extending work of Whittemore (1996), McPeek (1999) and Hossjer (2003d), the efficient score statistic is computed from a conditional likelihood of marker data given phenotypes. The resulting semiparametric linkage analysis is very similar to nonparametric linkage based on affected individuals. The efficient score S depends not only on identical-by-descent sharing and phenotypes, but also on a few parameters chosen by the user. We focus on (1) weak penetrance models, where the major gene has a small effect and (2) rare disease models, where the major gene has a possibly strong effect but the disease causing allele is rare. We illustrate our results for a large class of genetic models with a multivariate Gaussian liability. This class incorporates one major gene, polygenes and shared environmental effects in the liability, and allows e.g. binary, Gaussian, Poisson distributed and life-length phenotypes. A detailed simulation study is conducted for Gaussian phenotypes. The performance of the two optimal score functions S(wpairs) and S(normdom) are investigated. The conclusion is that (i) inclusion of polygenic effects into the score function increases overall performance for a wide range of genetic models and (ii) score functions based on the rare disease assumption are slightly more powerful.     

Polymyxin B-lipid interactions in Langmuir-Blodgett monolayers of Escherichia coli lipids: a thermodynamic and atomic force microscopy study

The dramatically increased frequency of antibiotic resistance has led to intensive efforts towards developing new families of antibiotics. Membrane-active antibiotic peptides such as polymyxin B (PxB) hold promise as the next generation of antibiotics, since they rarely spur the evolution of resistance. At low concentrations in the membrane, PxB forms vesicle-vesicle contacts and induces lipid exchange without leakage or fusion, a phenomenon that can explain its specificity towards gram-negative bacteria by contact formation between the two phospholipids interfaces in the periplasmatic space. In this work, the interaction of PxB and the nonantibiotic derivative polymyxin B nonapeptide (PxB-NP) with monolayers of Escherichia coli membrane lipids (ECL) has been studied by thermodynamic and structural methods. PxB inserts itself into ECL monolayers as a conformation that forms intermembrane contacts with vesicles injected underneath, and induces lipid exchange when the monolayer surface pressure is set at 32 mN/m (membrane equivalence pressure) or net transfer vesicle-to-monolayer at lower surface pressures. Thermodynamic analysis of the compression isotherms of mixed monolayers indicates that PxB inserts into the monolayer with an expansion of the mean molecular area, implying that peptide and lipids form nonideal mixtures. At low concentrations, corresponding to the membrane-membrane contact form of PxB, the mixed monolayers present positive excess energy values (deltaGm(Ex)), and atomic force microscopy (AFM) imaging reveals structures of approximately 120-nm diameter that protrude from the lipid surface approximately 0.7 nm. At concentrations of PxB above 4 mol %, thermodynamic analysis gives a very high deltaGm(Ex), corresponding to nonfavorable interactions, and AFM images show round structures of 20-30 nm diameter. PxB-NP behaves in a totally different way, in agreement with its inability to form vesicle-vesicle contacts and its lack of antibiotic effect. These results are discussed in the light of the mechanism of action of PxB on the membrane of gram-negative bacteria.     

Lipid-Membrane Affinity of Chimeric Metal-binding Green Fluorescent Protein

The Green Fluorescent Protein (GFP) is a useful marker to trace the expression of cellular proteins. However, little is known about changes in protein interaction properties after fusion to GFP. In this study, we present evidence for a binding affinity of chimeric cadmium-binding green fluorescent proteins to lipid membrane. This affinity has been observed in both cellular membranes and artificial lipid monolayers and bilayers. At the cellular level, the presence of Cd-binding peptide promoted the association of the chimeric GFP onto the lipid membrane, which declined the fluorescence emission of the engineered cells. Binding affinity to lipid membranes was further investigated using artificial lipid bilayers and monolayers. Small amounts of the chimeric GFP were found to incorporate into the lipid vesicles due to the high surface pressure of bilayer lipids. At low interfacial pressure of the lipid monolayer, incorporation of the chimeric Cd-binding GFP onto the lipid monolayer was revealed. From the measured lipid isotherms, we conclude that Cd-binding GFP mediates an increase in membrane fluidity and an expansion of the surface area of the lipid film. This evidence was strongly supported by epifluorescence microscopy, showing that the chimeric Cd-binding GFP preferentially binds to fluid-phase areas and defect parts of the lipid monolayer. All these findings demonstrate the hydrophobicity of the GFP constructs is mainly influenced by the fusion partner. Thus, the example of a metal-binding unit used here shines new light on the biophysical properties of GFP constructs.This revised version was published online in June 2005 with a corrected cover date.     

Growth of mixed cultures on mixtures of substitutable substrates: the operating diagram for a structured model

The growth of mixed microbial cultures on mixtures of substrates is a problem of fundamental biological interest. In the last two decades, several unstructured models of mixed-substrate growth have been studied. It is well known, however, that the growth patterns in mixed-substrate environments are dictated by the enzymes that catalyse the transport of substrates into the cell. We have shown previously that a model taking due account of transport enzymes captures and explains all the observed patterns of growth of a single species on two substitutable substrates (J. Theor. Biol. 190 (1998) 241). Here, we extend the model to study the steady states of growth of two species on two substitutable substrates. The model is analysed to determine the conditions for existence and stability of the various steady states. Simulations are performed to determine the flow rates and feed concentrations at which both species coexist. We show that if the interaction between the two species is purely competitive, then at any given flow rate, coexistence is possible only if the ratio of the two feed concentrations lies within a certain interval; excessive supply of either one of the two substrates leads to annihilation of one of the species. This result simplifies the construction of the operating diagram for purely competing species. This is because the two-dimensional surface that bounds the flow rates and feed concentrations at which both species coexist has a particularly simple geometry: It is completely determined by only two coordinates, the flow rate and the ratio of the two feed concentrations. We also study commensalistic interactions between the two species by assuming that one of the species excretes a product that can support the growth of the other species. We show that such interactions enhance the coexistence region.     

Biodiversity correlates with regional patterns of forest litter pools

We compared litter pools of more than 1,000 forests differing in tree species diversity over a large scale in Catalonia (NE Spain). Monospecific forests always had smaller litter pools than mixed (from 2 to 5 tree species) forests. Whether there was a positive effect beyond two species mixtures depended on the species and functional identity of the dominant tree species. In sclerophyllous forests the positive effect of diversity was a step-function from one to more species. However, in conifers, litter pools increased constantly with tree diversity. The identity of the dominant tree species and functional type had also a significant effect on litter pools. For instance, forests dominated by sclerophyllous tree species had larger litter pools than forests dominated by deciduous and conifer tree species. When other forest structure parameters (i.e. tree basal area, wood production, successional stage, shrub cover and leaf area index) and environmental factors (i.e. mean annual temperature, mean annual precipitation, annual evapotranspiration and hillside position) where included in the analysis only leaf area index, basal area, wood production and mean temperature influenced litter pools positively. Our analysis emphasizes that at the regional scale, the litter compartment can be as influenced by biodiversity components as by other forest structure and climate components. In mixed forests, species and functional identity of the trees determine whether litter pools increase with tree diversity.