TESTING STATISTICAL HYPOTHESIS ON RANDOM TREES AND APPLICATIONS TO THE PROTEIN CLASSIFICATION PROBLEM

Efficient automatic protein classification is of central importance in genomic annotation. As an independent way to check the reliability of the classification, we propose a statistical approach to test if two sets of protein domain sequences coming from two families of the Pfam database are significantly different. We model protein sequences as realizations of Variable Length Markov Chains (VLMC) and we use the context trees as a signature of each protein family. Our approach is based on a Kolmogorov-Smirnov-type goodness-of-fit test proposed by Balding et at. [Limit theorems for sequences of random trees (2008), DOI: 10.1007/s11749-008-0092-z]. The test statistic is a supremum over the space of trees of a function of the two samples; its computation grows, in principle, exponentially fast with the maximal number of nodes of the potential trees. We show how to transform this problem into a max-flow over a related graph which can be solved using a Ford-Fulkerson algorithm in polynomial time on that number. We apply the test to 10 randomly chosen protein domain families from the seed of Pfam-A database (high quality, manually curated families). The test shows that the distributions of context trees coming from different families are significantly different. We emphasize that this is a novel mathematical approach to validate the automatic clustering of sequences in any context. We also study the performance of the test via simulations on Galton-Watson related processes.

Survival time of random walk in random environment among soft obstacles

We consider a Random Walk in Random Environment (RWRE) moving in an i.i.d. random field of obstacles. When the particle hits an obstacle, it disappears with a positive probability. We obtain quenched and annealed bounds on the tails of the survival time in the general d-dimensional case. We then consider a simplified one-dimensional model (where transition probabilities and obstacles are independent and the RWRE only moves to neighbour sites), and obtain finer results for the tail of the survival time. In addition, we study also the ""mixed"" probability measures (quenched with respect to the obstacles and annealed with respect to the transition probabilities and vice-versa) and give results for tails of the survival time with respect to these probability measures. Further, we apply the same methods to obtain bounds for the tails of hitting times of Branching Random Walks in Random Environment (BRWRE).

Random perturbations of stochastic processes with unbounded variable length memory

We consider binary infinite order stochastic chains perturbed by a random noise. This means that at each time step, the value assumed by the chain can be randomly and independently flipped with a small fixed probability. We show that the transition probabilities of the perturbed chain are uniformly close to the corresponding transition probabilities of the original chain. As a consequence, in the case of stochastic chains with unbounded but otherwise finite variable length memory, we show that it is possible to recover the context tree of the original chain, using a suitable version of the algorithm Context, provided that the noise is small enough.

Gibbs random graphs on point processes

Consider a discrete locally finite subset Gamma of R(d) and the cornplete graph (Gamma, E), with vertices Gamma and edges E. We consider Gibbs measures on the set of sub-graphs with vertices Gamma and edges E` subset of E. The Gibbs interaction acts between open edges having a vertex in common. We study percolation properties of the Gibbs distribution of the graph ensemble. The main results concern percolation properties of the open edges in two cases: (a) when Gamma is sampled from a homogeneous Poisson process; and (b) for a fixed Gamma with sufficiently sparse points. (c) 2010 American Institute of Physics. [doi:10.1063/1.3514605]

An oracle approach for interaction neighborhood estimation in random fields

We consider the problem of interaction neighborhood estimation from the partial observation of a finite number of realizations of a random field. We introduce a model selection rule to choose estimators of conditional probabilities among natural candidates. Our main result is an oracle inequality satisfied by the resulting estimator. We use then this selection rule in a two-step procedure to evaluate the interacting neighborhoods. The selection rule selects a small prior set of possible interacting points and a cutting step remove from this prior set the irrelevant points. We also prove that the Ising models satisfy the assumptions of the main theorems, without restrictions on the temperature, on the structure of the interacting graph or on the range of the interactions. It provides therefore a large class of applications for our results. We give a computationally efficient procedure in these models. We finally show the practical efficiency of our approach in a simulation study.

Surface effects in the magnetic properties of crystalline 3 nm ferrite nanoparticles chemically synthesized

We have systematically studied the magnetic properties of ferrite nanoparticles with 3, 7, and 11 nm of diameter with very narrow grain size distributions. Samples were prepared by the thermal decomposition of Fe (acac)(3) in the presence of surfactants giving nanoparticles covered by oleic acid. High resolution transmission electron microscopy (HRTEM) images and XRD diffraction patterns confirms that all samples are composed by crystalline nanoparticles with the spinel structure expected for the iron ferrite. ac and dc magnetization measurements, as well in-field Mossbauer spectroscopy, indicate that the magnetic properties of nanoparticles with 11 and 7 nm are close to those expected for a monodomain, presenting large M(S) (close to the magnetite bulk). Despite the crystalline structure observed in HRTEM images, the nanoparticles with 3 nm are composed by a magnetically ordered region (core) and a surface region that presents a different magnetic order and it contains about 66% of Fe atoms. The high saturation and irreversibility fields in the M(H) loops of the particles with 3 nm together with the misalignment at 120 kOe in the in-field Mossbauer spectrum of surface component indicate a high surface anisotropy for the surface atoms, which is not observed for the core. For T < 10 K, we observe an increase in the susceptibility and of the magnetization for former sample, indicating that surface moments tend to align with applied field increasing the magnetic core size. (C) 2010 American Institute of Physics. [doi:10.1063/1.3514585]

Full-dimensional analytical ab initio potential energy surface of the ground state of HOI

Extensive ab initio calculations using a complete active space second-order perturbation theory wavefunction, including scalar and spin-orbit relativistic effects with a quadruple-zeta quality basis set were used to construct an analytical potential energy surface (PES) of the ground state of the [H, O, I] system. A total of 5344 points were fit to a three-dimensional function of the internuclear distances, with a global root-mean-square error of 1.26 kcal mol(-1). The resulting PES describes accurately the main features of this system: the HOI and HIO isomers, the transition state between them, and all dissociation asymptotes. After a small adjustment, using a scaling factor on the internal coordinates of HOI, the frequencies calculated in this work agree with the experimental data available within 10 cm(-1). (C) 2011 American Institute of Physics. [doi: 10.1063/1.3615545]

High performance gold nanorods and silver nanocubes in surface-enhanced Raman spectroscopy of pesticides

The behavior of Au nanorods and Ag nanocubes as analytical sensors was evaluated for three different classes of herbicides. The use of such anisotropic nanoparticles in surface-enhanced Raman scattering (SERS) experiments allows the one to obtain the spectrum of crystal violet dye in the single molecule regime, as well as the pesticides dichlorophenoxyacetic acid (2,4-D), trichlorfon and ametryn. Such metallic substrates show high SERS performance at low analyte concentrations making them adequate for use as analytical sensors. Density functional theory (DFT) calculations of the geometries and vibrational wavenumbers of the adsorbates in the presence of silver or gold atoms were used to elucidate the nature of adsorbate-nanostructure bonding in each case and support the enhancement patterns observed in each SERS spectrum.

Chromic materials for responsive surface-enhanced resonance Raman scattering systems: a nanometric pH sensor

The use of chromic materials for responsive surface-enhanced resonance Raman scattering (SERRS) based nanosensors is reported. The potential of nano-chromic SERRS is demonstrated with the use of the halochrome methyl yellow to fabricate an ultrasensitive pH optical sensor. Some of the challenges of the incorporation of chromic materials with metal nanostructures are addressed through the use of computational calculations and a comparison to measured SERRS and surface-enhanced Raman scattering (SERS) spectra is presented. A strong correlation between the measured SERRS and the medium's proton concentration is demonstrated for the pH range 2-6. The high sensitivity achieved by the use of resonance Raman conditions is shown through responsive SERRS measurements from only femtolitres of volume and with the concentration of the reporting molecules approaching the single molecule regime.

Surface structure effects on the electrochemical oxidation of ethanol on platinum single crystal electrodes

Ethanol oxidation has been studied on Pt(111), Pt(100) and Pt(110) electrodes in order to investigate the effect of the surface structure and adsorbing anions using electrochemical and FTIR techniques. The results indicate that the surface structure and anion adsorption affect significantly the reactivity of the electrode. Thus, the main product of the oxidation of ethanol on the Pt(111) electrode is acetic acid, and acetaldehyde is formed as secondary product. Moreover, the amount of CO formed is very small, and probably associated with the defects present on the electrode surface. For that reason, the amount of CO(2) is also small. This electrode has the highest catalytic activity for the formation of acetic acid in perchloric acid. However, the formation of acetic acid is inhibited by the presence of specifically adsorbed anions, such as (bi) sulfate or acetate, which is the result of the formation of acetic acid. On the other hand, CO is readily formed at low potentials on the Pt(100) electrode, blocking completely the surface. Between 0.65 and 0.80 V, the CO layer is oxidized and the production of acetaldehyde and acetic acid is detected. The Pt(110) electrode displays the highest catalytic activity for the splitting of the C-C bond. Reactions giving rise to CO formation, from either ethanol or acetaldehyde, occur at high rate at any potential. On the other hand, the oxidation of acetaldehyde to acetic acid has probably the lower reaction rate of the three basal planes.

Reconstruction of core and surface nanoparticles: The example of Pt(55) and Au(55)

Cuboctahedron (CUB) and icosahedron (ICO) model structures are widely used in the study of transition-metal (TM) nanoparticles (NPs), however, it might not provide a reliable description for small TM NPs such as the Pt(55) and Au(55) systems in gas phase. In this work, we combined density-functional theory calculations with atomic configurations generated by the basin hopping Monte Carlo algorithm within the empirical Sutton-Chen embedded atom potential. We identified alternative lower energy configurations compared with the ICO and CUB model structures, e. g., our lowest energy structures are 5.22 eV (Pt(55)) and 2.01 eV (Au(55)) lower than ICO. The energy gain is obtained by the Pt and Au diffusion from the ICO core region to the NP surface, which is driven by surface compression (only 12 atoms) on the ICO core region. Therefore, in the lowest energy configurations, the core size reduces from 13 atoms (ICO, CUB) to about 9 atoms while the NP surface increases from 42 atoms (ICO, CUB) to about 46 atoms. The present mechanism can provide an improved atom-level understanding of small TM NPs reconstructions.

Nanogravimetric and voltammetric studies of a Pt-Rh alloy surface and its behavior for methanol oxidation

This paper describes the preparation of a Pt-Rh alloy surface electrodeposited on Pt electrodes and its electrocatalytic characterization for methanol oxidation. The X-ray photoelectronic spectroscopy ( XPS) results demonstrate that the surface composition is approximately 24 at-% Rh and 76 % Pt. The cyclic voltammetry (CV) and electrochemical quartz crystal (EQCN) results for the alloy were associated, for platinum, to the well known profile in acidic medium. For Rh, on the alloy, the generation of rhodium hydroxide species (Rh(OH)(3) and RhO(OH)(3)) was measured. During the successive oxidation-reduction cycles the mass returns to its original value, indicating the reversibility of the processes. It was not observed rhodium dissolution during the cycling. The 76/24 at % Pt-Rh alloy presented singular electrocatalytic activity for methanol electrooxidation, which started at more negative potentials compared to pure Pt (70 mV). During the sweep towards more negative potentials, there is only weak CO re-adsorption on both Rh and Pt-Rh alloy surfaces, which can be explained by considering the interaction energy between Rh and CO.

Influence of land-use change on near-surface hydrological processes: Undisturbed forest to pasture

Soil compaction that follows the clearing of tropical forest for cattle pasture is associated with lower soil hydraulic conductivity and increased frequency and volume of overland flow. We investigated the frequency of perched water tables, overland flow and stormflow in an Amazon forest and in an adjacent 25-year-old pasture cleared from the same forest. We compared the results with the frequencies of these phenomena estimated from comparisons of rainfall intensity and soil hydraulic conductivity. The frequency of perched water tables based on rainfall intensity and soil hydraulic conductivity was expected to double in pasture compared with forest. This corresponded closely with an approximate doubling of the frequency of stormflow and overland flow in pasture. In contrast, the stormflow volume in pasture increased 17-fold. This disproportional increase of stormflow resulted from overland flow generation over large areas of pasture, while overland flow generation in the forest was spatially limited and was observed only very near the stream channel. In both catchments, stormflow was generated by saturation excess because of perched water tables and near-surface groundwater levels. Stormflow was occasionally generated in the forest by rapid return flow from macropores, while slow return flow from a continuous perched water table was more common in the pasture. These results suggest that deforestation for pasture alters fundamental mechanisms of stormflow generation and may increase runoff volumes over wide regions of Amazonia. (C) 2009 Elsevier B.V. All rights reserved.

Relevance of leaf surface contamination for assessing chemical composition of bromeliads in a restinga forest

Resuspended soil and other airborne particles adhered to the leaf surface affect the chemical composition of the plant. A well-defined cleaning procedure is necessary to avoid this problem, providing a correct assessment of the inherent chemical composition of bromeliads. To evaluate the influence of a washing procedure, INAA was applied for determining chemical elements in the leaves of bromeliads from Vriesea carinata species, both non-washed and washed with Alconox, EDTA and bi-distilled water. Br, Ce, Hg, La, Sc, Se, Sm and Th showed higher mass fractions in non-washed leaves. The washing procedure removed the exogenous material without leaching chemical elements from inside the tissues.

Middle Pleistocene sea surface temperature in the Brazil-Malvinas Confluence Zone: Paleoceanographic implications based on planktonic foraminifera

We reconstructed Middle Pleistocene surface hydrography in the western South Atlantic based on planktonic foraminiferal assemblages, modern analog technique and Globorotalia truncatulinoides isotopic ratios of core SP1251 (38 degrees 29.7`S / 53 degrees 40.7`W / 3400 m water depth). Biostratigraphic analysis suggests that sediments were deposited between 0.3 and 0.12 Ma and therefore correlate to Marine Isotopic Stage 6 or 8. Faunal assemblage-based winter and summer SST estimates suggest that the western South Atlantic at 38 degrees S was 4-6 degrees C colder than at present, within the expected range for a glacial interval. High relative abundances of subantarctic species, particularly the dominance of Neogloboquadrina pachyderma (left), support lower than present SSTs throughout the recorded period. The oxygen isotopic composition of G. truncatulinoides suggests a northward shift of the Brazil-Malvinas Confluence Zone and of the associated mid-latitude frontal system during this Middle Pleistocene cold period, and a stronger than present influence of superficial subantarctic waters and lowering in SSTs at our core site during the recorded Middle Pleistocene glacial.