Likelihood-free estimation of model evidence

Statistical methods of inference typically require the likelihood function to be computable in a reasonable amount of time. The class of “likelihood-free” methods termed Approximate Bayesian Computation (ABC) is able to eliminate this requirement, replacing the evaluation of the likelihood with simulation from it. Likelihood-free methods have gained in efficiency and popularity in the past few years, following their integration with Markov Chain Monte Carlo (MCMC) and Sequential Monte Carlo (SMC) in order to better explore the parameter space. They have been applied primarily to estimating the parameters of a given model, but can also be used to compare models. Here we present novel likelihood-free approaches to model comparison, based upon the independent estimation of the evidence of each model under study. Key advantages of these approaches over previous techniques are that they allow the exploitation of MCMC or SMC algorithms for exploring the parameter space, and that they do not require a sampler able to mix between models. We validate the proposed methods using a simple exponential family problem before providing a realistic problem from human population genetics: the comparison of different demographic models based upon genetic data from the Y chromosome.

Open solar flux estimates from near-Earth measurements of the interplanetary magnetic field: comparison of the first two perihelion passes of the Ulysses spacecraft

Results from all phases of the orbits of the Ulysses spacecraft have shown that the magnitude of the radial component of the heliospheric field is approximately independent of heliographic latitude. This result allows the use of near- Earth observations to compute the total open flux of the Sun. For example, using satellite observations of the interplanetary magnetic field, the average open solar flux was shown to have risen by 29% between 1963 and 1987 and using the aa geomagnetic index it was found to have doubled during the 20th century. It is therefore important to assess fully the accuracy of the result and to check that it applies to all phases of the solar cycle. The first perihelion pass of the Ulysses spacecraft was close to sunspot minimum, and recent data from the second perihelion pass show that the result also holds at solar maximum. The high level of correlation between the open flux derived from the various methods strongly supports the Ulysses discovery that the radial field component is independent of latitude. We show here that the errors introduced into open solar flux estimates by assuming that the heliospheric field’s radial component is independent of latitude are similar for the two passes and are of order 25% for daily values, falling to 5% for averaging timescales of 27 days or greater. We compare here the results of four methods for estimating the open solar flux with results from the first and second perehelion passes by Ulysses. We find that the errors are lowest (1–5% for averages over the entire perehelion passes lasting near 320 days), for near-Earth methods, based on either interplanetary magnetic field observations or the aa geomagnetic activity index. The corresponding errors for the Solanki et al. (2000) model are of the order of 9–15% and for the PFSS method, based on solar magnetograms, are of the order of 13–47%. The model of Solanki et al. is based on the continuity equation of open flux, and uses the sunspot number to quantify the rate of open flux emergence. It predicts that the average open solar flux has been decreasing since 1987, as Correspondence to: M. Lockwood (m.lockwood@rl.ac.uk) is observed in the variation of all the estimates of the open flux. This decline combines with the solar cycle variation to produce an open flux during the second (sunspot maximum) perihelion pass of Ulysses which is only slightly larger than that during the first (sunspot minimum) perihelion pass.

Military targeting in the context of self-defence actions

In order to bring conceptual clarity to a particular dimension of the relationship between the jus ad bellum and the jus in bello regimes, this article explores the independent sources of a military targeting rule in both branches of international law. The aim is not to displace the jus in bello as the ‘lead’ regime on how targeting decisions must be made, or to undermine the traditional separation between these regimes. Rather, conceptual light is shed on a sometimes assumed but generally neglected dimension of the jus ad bellum’s necessity and proportionality criteria that may, in limited circumstances, have significance for our understanding of human protection during war, by covering possible gaps in the jus in bello targeting rules.

Parametrizing the horizontal inhomogeneity of ice water content using CloudSat data products

In order to calculate unbiased microphysical and radiative quantities in the presence of a cloud, it is necessary to know not only the mean water content but also the distribution of this water content. This article describes a study of the in-cloud horizontal inhomogeneity of ice water content, based on CloudSat data. In particular, by focusing on the relations with variables that are already available in general circulation models (GCMs), a parametrization of inhomogeneity that is suitable for inclusion in GCM simulations is developed. Inhomogeneity is defined in terms of the fractional standard deviation (FSD), which is given by the standard deviation divided by the mean. The FSD of ice water content is found to increase with the horizontal scale over which it is calculated and also with the thickness of the layer. The connection to cloud fraction is more complicated; for small cloud fractions FSD increases as cloud fraction increases while FSD decreases sharply for overcast scenes. The relations to horizontal scale, layer thickness and cloud fraction are parametrized in a relatively simple equation. The performance of this parametrization is tested on an independent set of CloudSat data. The parametrization is shown to be a significant improvement on the assumption of a single-valued global FSD

DENSITY PROFILE, VELOCITY ANISOTROPY, AND LINE-OF-SIGHT EXTERNAL CONVERGENCE OF SLACS GRAVITATIONAL LENSES

Data from 58 strong-lensing events surveyed by the Sloan Lens ACS Survey are used to estimate the projected galaxy mass inside their Einstein radii by two independent methods: stellar dynamics and strong gravitational lensing. We perform a joint analysis of these two estimates inside models with up to three degrees of freedom with respect to the lens density profile, stellar velocity anisotropy, and line-of-sight (LOS) external convergence, which incorporates the effect of the large-scale structure on strong lensing. A Bayesian analysis is employed to estimate the model parameters, evaluate their significance, and compare models. We find that the data favor Jaffe`s light profile over Hernquist`s, but that any particular choice between these two does not change the qualitative conclusions with respect to the features of the system that we investigate. The density profile is compatible with an isothermal, being sightly steeper and having an uncertainty in the logarithmic slope of the order of 5% in models that take into account a prior ignorance on anisotropy and external convergence. We identify a considerable degeneracy between the density profile slope and the anisotropy parameter, which largely increases the uncertainties in the estimates of these parameters, but we encounter no evidence in favor of an anisotropic velocity distribution on average for the whole sample. An LOS external convergence following a prior probability distribution given by cosmology has a small effect on the estimation of the lens density profile, but can increase the dispersion of its value by nearly 40%.

Multilocus Analyses of Seven Candidate Genes Suggest Interacting Pathways for Obesity-Related Traits in Brazilian Populations

We investigated whether variants in major candidate genes for food intake and body weight regulation contribute to obesity-related traits under a multilocus perspective. We studied 375 Brazilian subjects from partially isolated African-derived populations (quilombos). Seven variants displaying conflicting results in previous reports and supposedly implicated in the susceptibility of obesity-related phenotypes were investigated: beta(2)-adrenergic receptor (ADRB2) (Arg16Gly), insulin induced gene 2 (INSIG2) (rs7566605), leptin (LEP) (A19G), LEP receptor (LEPR) (Gln223Arg), perilipin (PLIN) (6209T > C), peroxisome proliferator-activated receptor-gamma (PPARG) (Pro12Ala), and resistin (RETN) (-420C > G). Regression models as well as generalized multifactor dimensionality reduction (GMDR) were employed to test the contribution of individual effects and higher-order interactions to BMI and waist-hip ratio (WHR) variation and risk of overweight/obesity. The best multilocus association signal identified in the quilombos was further examined in an independent sample of 334 Brazilian subjects of European ancestry. In quilombos, only the PPARG polymorphism displayed significant individual effects (WHR variation, P = 0.028). No association was observed either with the risk of overweight/obesity (BMI >= 25 kg/m(2)), risk of obesity alone (BMI >= 30 kg/m(2)) or BMI variation. However, GMDR analyses revealed an interaction between the LEPR and ADRB2 polymorphisms (P = 0.009) as well as a third-order effect involving the latter two variants plus INSIG2 (P = 0.034) with overweight/obesity. Assessment of the LEPR-ADRB2 interaction in the second sample indicated a marginally significant association (P = 0.0724), which was further verified to be limited to men (P = 0.0118). Together, our findings suggest evidence for a two-locus interaction between the LEPR Gln223Arg and ADRB2 Arg16Gly variants in the risk of overweight/obesity, and highlight further the importance of multilocus effects in the genetic component of obesity.

First order reversal curve analysis of nanocrystalline Pd(80)Co(20) alloy films

Films of isotropic nanocrystalline Pd(80)Co(20) alloys were obtained by electrodeposition onto brass substrate in plating baths maintained at different pH values. Increasing the pH of the plating bath led to an increase in mean grain size without inducing significant changes in the composition of the alloy. The magnetocrystalline anisotropy constant was estimated and the value was of the same order of magnitude as that reported for samples with perpendicular magnetic anisotropy. First order reversal curve (FORC) analysis revealed the presence of an important component of reversible magnetization. Also, FORC diagrams obtained at different sweep rate of the applied magnetic field, revealed that this reversible component is strongly affected by kinetic effect. The slight bias observed in the irreversible part of the FORC distribution suggested the dominance of magnetizing intergrain exchange coupling over demagnetizing dipolar interactions and microstructural disorder. (c) 2009 Elsevier B.V. All rights reserved.

Solvation in Pure Liquids: What Can Be Learned from the Use of Pairs of Indicators?

The solvation of six solvatochromic probes in a large number of solvents (33-68) was examined at 25 degrees C. The probes employed were the following: 2,6-diphenyl-4-(2,4,6-triphenylpyridinium-1-yl) phenolate (RB); 4-[(E)2-(1-methylpyridinium-4-yl)ethenyl] phenolate, MePM; 1-methylquinolinium-8-olate, QB; 2-bromo-4-[(E)-2-(1-methylpyridinium-4-yl)ethenyl] phenolate, MePMBr, 2,6-dichloro-4-(2,4,6-triphenyl pyridinium-1-yl) phenolate (WB); and 2,6-dibromo-4-[(E)-2-(1-methylpyridinium-4-yl)ethenyl] phenolate, MePMBr(2), respectively. Of these, MePMBr is a novel compound. They can be grouped in three pairs, each with similar pK(a) in water but with different molecular properties, for example, lipophilicity and dipole moment. These pairs are formed by RB and MePM; QB and MePMBr; WB and MePMBr(2), respectively. Theoretical calculations were carried out in order to calculate their physicochemical properties including bond lengths, dihedral angles, dipole moments, and wavelength of absorption of the intramolecular charge-transfer band in four solvents, water, methanol, acetone, and DMSO, respectively. The data calculated were in excellent agreement with available experimental data, for example, bond length and dihedral angles. This gives credence to the use of the calculated properties in explaining the solvatochromic behaviors observed. The dependence of an empirical solvent polarity scale E(T)(probe) in kcal/mol on the physicochemical properties of the solvent (acidity, basicity, and dipolarity/polarizability) and those of the probes (pK(a), and dipole moment) was analyzed by using known multiparameter solvation equations. For each pair of probes, values of E(T)(probe) (for example, E(T)(MePM) versus E(T)(RB)) were found to be linearly correlated with correlation coefficients, r, between 0.9548 and 0.9860. For the mercyanine series, the values of E(T)(probe) also correlated linearly, with (r) of 0.9772 (MePMBr versus MePM) and 0.9919 (MePMBr(2) versus MePM). The response of each pair of probes (of similar pK(a)) to solvent acidity is the same, provided that solute-solvent hydrogen-bonding is not seriously affected by steric crowding (as in case of RB). We show, for the first time, that the response to solvent dipolarity/polarizability is linearly correlated to the dipole moment of the probes. The successive introduction of bromine atoms in MePM (to give MePMBr, then MePMBr(2)) leads to the following linear decrease: pK(a) in water, length of the phenolate oxygen-carbon bond, length of the central ethylenic bond, susceptibility to solvent acidity, and susceptibility to solvent dipolarity/polarizability. Thus studying the solvation of probes whose molecular structures are varied systematically produces a wealth of information on the effect of solute structure on its solvation. The results of solvation of the present probes were employed in order to test the goodness of fit of two independent sets of solvent solvatochromic parameters.

Micellar properties of surface active ionic liquids: A comparison of 1-hexadecyl-3-methylimidazolium chloride with structurally related cationic surfactants

Ionic liquids, ILs, carrying long-chain alkyl groups are surface active, SAIIs. We investigated the micellar properties of the SAIL 1-hexadecyl-3-methylimidazolium chloride, C(16)MeImCl, and compared the data with 1-hexadecylpyridinium chloride, C(16)PYCl, and benzyl (3-hexadecanoylaminoethyl)dimethylammonium chloride, C(15)AEtBzMe(2)Cl. The properties compared include critical micelle concentration, cmc; thermodynamic parameters of micellization; empirical polarity and water concentrations in the interfacial regions. In the temperature range from 15 to 75 degrees C, the order of cmc in H(2)O and in D(2)O is C(16)PYCl > C(16)MeImCl > C(15)AEtBzMe(2)Cl. The enthalpies of micellization, Delta H(mic)(degrees), were calculated indirectly from by use of the van`t Hoff treatment; directly by isothermal titration calorimetry, ITC. Calculation of the degree of counter-ion dissociation, alpha(mic), from conductivity measurements, by use of Evans equation requires knowledge of the aggregation numbers, N(agg), at different temperatures. We have introduced a reliable method for carrying out this calculation, based on the volume and length of the monomer, and the dependence of N(agg) on temperature. The N(agg) calculated for C(16)PyCl and C(16)MeImCl were corroborated by light scattering measurements. Conductivity- and ITC-based Delta H(mic)(degrees) do not agree; reasons for this discrepancy are discussed. Micelle formation is entropy driven: at all studied temperatures for C(16)MeImCl; only up to 65 degrees C for C(16)PyCl; and up to 55 degrees C for C(15)AEtBzMe(2)Cl. All these data can be rationalized by considering hydrogen-bonding between the head-ions of the monomers in the micellar aggregate. The empirical polarities and concentrations of interfacial water were found to be independent of the nature of the head-group. (C) 2010 Elsevier Inc. All rights reserved.

On the temperature stability of extracellular hemoglobin of Glossoscolex paulistus, at different oxidation states: SAXS and DLS studies

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

New constraints on ages of glasses proposed as reference materials for fission-track dating

New analyses have been performed in order to enhance the data-set on the independent ages of four glasses that have been proposed as reference materials for fission-track dating. The results are as follows. Moldavite - repeated (40)Ar/(39)Ar age determinations on samples from deposits from Bohemia and Moravia yielded an average of 14.34 +/- 0.08 Ma. This datum agrees with other recent determinations and is significantly younger than the (40)Ar/(39)Ar age of 15.21 +/- 0.15 Ma determined in the early 1980s. Macusanite (Peru) -four K-Ar ages ranging from 5.44 +/- 0.06 to 5.72 +/- 0.12 Ma have been published previously. New (40)Ar/(39)Ar ages gave an average of 5.12 +/- 0.04 Ma. Plateau fission-track ages determined using the IRMM-540 certified glass and U and Th thin films for neutron fluence measurements agree better with these new (40)Ar/(39)Ar ages than the previously published ages. Roccastrada glass (Italy) - a new (40)Ar/(39)Ar age, 2.45 +/- 0.04 Ma, is consistent with previous determinations. The Quiron obsidian (Argentina) is a recently discovered glass that has been proposed as an additional reference material for its high spontaneous track density (around 100 000 cm(-2)). Defects that might produce spurious tracks are virtually absent. An independent (40)Ar/(39)Ar age of 8.77 +/- 0.09 Ma was determined and is recommended for this glass. We believe that these materials, which will be distributed upon request to fission-track groups, will be very useful for testing system calibrations and experimental procedures.

Stress induces behavioral sensitization, increases nicotine-seeking behavior and leads to a decrease of CREB in the nucleus accumbens

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Activation of Ca2+-independent membrane-damaging activity in Lys49-phospholipase A(2) promoted by amphiphilic molecules

Association of class-II phospholipase A(2) (PLA(2)) with aggregated phospholipid substrate results in elevated levels of the Ca2+-dependent hydrolytic activity. The Asp49 residue participates in coordination of the Ca2+ ion cofactor, however, in Lys49-PLA(2) homologues (Lys49-PLA(2)S), substitution of the Asp49 by Lys results in loss of Ca2+ binding and lack of detectable phospholipid hydrolysis. Nevertheless, Lys49-PLA2S cause Ca2+-independent damage of liposome membranes. Bothropstoxin-I is a homodimeric Lys49-PLA(2) from the venom of Bothrops jararacussu, and in fluorescent marker release and dynamic light scattering experiments with DPPC liposomes we demonstrate activation of the Ca2+-independent membrane damaging activity by similar to4 molecules of sodium dodecyl sulphate (SDS) per protein monomer. Activation is accomparlied by significant changes in the intrinsic tryptophan fluorescence emission (ITFE) and near UV circular dichroism (UVCD) spectra of the protein. Subsequent binding of 7-10 SDS molecules results in further alterations in the ITFE and far UVCD spectra. Reduction in the rate of N-bromosuccinimide modification of Trp77 at the dimer interface suggests that initial binding of SDS to this region accompanies the activation of the membrane damaging activity. 1-anilinonaphthalene-8-sulphonic acid binding studies indicate that subsequent SDS binding to the active site is concomitant with the second structural transition. These results provide insights in the structural basis of amphiphile/protein coupling in class-II PLA(2)s. (C) 2004 Published by Elsevier B.V.

Shape, transverse size, and charged-hadron multiplicity of jets in pp collisions at root s=7 TeV

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

First order reversal curve analysis of nanocrystalline Pd80Co20 alloy films

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)