Étude de la performance d’un algorithme Metropolis-Hastings avec ajustement directionnel

Les méthodes de Monte Carlo par chaîne de Markov (MCMC) sont des outils très populaires pour l’échantillonnage de lois de probabilité complexes et/ou en grandes dimensions. Étant donné leur facilité d’application, ces méthodes sont largement répandues dans plusieurs communautés scientifiques et bien certainement en statistique, particulièrement en analyse bayésienne. Depuis l’apparition de la première méthode MCMC en 1953, le nombre de ces algorithmes a considérablement augmenté et ce sujet continue d’être une aire de recherche active. Un nouvel algorithme MCMC avec ajustement directionnel a été récemment développé par Bédard et al. (IJSS, 9 :2008) et certaines de ses propriétés restent partiellement méconnues. L’objectif de ce mémoire est de tenter d’établir l’impact d’un paramètre clé de cette méthode sur la performance globale de l’approche. Un second objectif est de comparer cet algorithme à d’autres méthodes MCMC plus versatiles afin de juger de sa performance de façon relative.

Recyclage des candidats dans l'algorithme Metropolis à essais multiples

Les méthodes de Monte Carlo par chaînes de Markov (MCCM) sont des méthodes servant à échantillonner à partir de distributions de probabilité. Ces techniques se basent sur le parcours de chaînes de Markov ayant pour lois stationnaires les distributions à échantillonner. Étant donné leur facilité d’application, elles constituent une des approches les plus utilisées dans la communauté statistique, et tout particulièrement en analyse bayésienne. Ce sont des outils très populaires pour l’échantillonnage de lois de probabilité complexes et/ou en grandes dimensions. Depuis l’apparition de la première méthode MCCM en 1953 (la méthode de Metropolis, voir [10]), l’intérêt pour ces méthodes, ainsi que l’éventail d’algorithmes disponibles ne cessent de s’accroître d’une année à l’autre. Bien que l’algorithme Metropolis-Hastings (voir [8]) puisse être considéré comme l’un des algorithmes de Monte Carlo par chaînes de Markov les plus généraux, il est aussi l’un des plus simples à comprendre et à expliquer, ce qui en fait un algorithme idéal pour débuter. Il a été sujet de développement par plusieurs chercheurs. L’algorithme Metropolis à essais multiples (MTM), introduit dans la littérature statistique par [9], est considéré comme un développement intéressant dans ce domaine, mais malheureusement son implémentation est très coûteuse (en termes de temps). Récemment, un nouvel algorithme a été développé par [1]. Il s’agit de l’algorithme Metropolis à essais multiples revisité (MTM revisité), qui définit la méthode MTM standard mentionnée précédemment dans le cadre de l’algorithme Metropolis-Hastings sur un espace étendu. L’objectif de ce travail est, en premier lieu, de présenter les méthodes MCCM, et par la suite d’étudier et d’analyser les algorithmes Metropolis-Hastings ainsi que le MTM standard afin de permettre aux lecteurs une meilleure compréhension de l’implémentation de ces méthodes. Un deuxième objectif est d’étudier les perspectives ainsi que les inconvénients de l’algorithme MTM revisité afin de voir s’il répond aux attentes de la communauté statistique. Enfin, nous tentons de combattre le problème de sédentarité de l’algorithme MTM revisité, ce qui donne lieu à un tout nouvel algorithme. Ce nouvel algorithme performe bien lorsque le nombre de candidats générés à chaque itérations est petit, mais sa performance se dégrade à mesure que ce nombre de candidats croît.

Experimental verification of suction sampler capture efficiency in grasslands of differing vegetation height and structure

1. Suction sampling is a popular method for the collection of quantitative data on grassland invertebrate populations, although there have been no detailed studies into the effectiveness of the method. 2. We investigate the effect of effort (duration and number of suction samples) and sward height on the efficiency of suction sampling of grassland beetle, true bug, planthopper and spider Populations. We also compare Suction sampling with an absolute sampling method based on the destructive removal of turfs. 3. Sampling for durations of 16 seconds was sufficient to collect 90% of all individuals and species of grassland beetles, with less time required for the true bugs, spiders and planthoppers. The number of samples required to collect 90% of the species was more variable, although in general 55 sub-samples was sufficient for all groups, except the true bugs. Increasing sward height had a negative effect on the capture efficiency of suction sampling. 4. The assemblage structure of beetles, planthoppers and spiders was independent of the sampling method (suction or absolute) used. 5. Synthesis and applications. In contrast to other sampling methods used in grassland habitats (e.g. sweep netting or pitfall trapping), suction sampling is an effective quantitative tool for the measurement of invertebrate diversity and assemblage structure providing sward height is included as a covariate. The effective sampling of beetles, true bugs, planthoppers and spiders altogether requires a minimum sampling effort of 110 sub-samples of duration of 16 seconds. Such sampling intensities can be adjusted depending on the taxa sampled, and we provide information to minimize sampling problems associated with this versatile technique. Suction sampling should remain an important component in the toolbox of experimental techniques used during both experimental and management sampling regimes within agroecosystems, grasslands or other low-lying vegetation types.

Using data from an encounter sampler to model fish dispersal

A method to estimate speed of free-ranging fishes using a passive sampling device is described and illustrated with data from the Everglades, U.S.A. Catch per unit effort (CPUE) from minnow traps embedded in drift fences was treated as an encounter rate and used to estimate speed, when combined with an independent estimate of density obtained by use of throw traps that enclose 1 m2 of marsh habitat. Underwater video was used to evaluate capture efficiency and species-specific bias of minnow traps and two sampling studies were used to estimate trap saturation and diel-movement patterns; these results were used to optimize sampling and derive correction factors to adjust species-specific encounter rates for bias and capture efficiency. Sailfin mollies Poecilia latipinna displayed a high frequency of escape from traps, whereas eastern mosquitofish Gambusia holbrooki were most likely to avoid a trap once they encountered it; dollar sunfish Lepomis marginatus were least likely to avoid the trap once they encountered it or to escape once they were captured. Length of sampling and time of day affected CPUE; fishes generally had a very low retention rate over a 24 h sample time and only the Everglades pygmy sunfish Elassoma evergladei were commonly captured at night. Dispersal speed of fishes in the Florida Everglades, U.S.A., was shown to vary seasonally and among species, ranging from 0· 05 to 0· 15 m s−1 for small poeciliids and fundulids to 0· 1 to 1· 8 m s−1 for L. marginatus. Speed was generally highest late in the wet season and lowest in the dry season, possibly tied to dispersal behaviours linked to finding and remaining in dry-season refuges. These speed estimates can be used to estimate the diffusive movement rate, which is commonly employed in spatial ecological models.

Time-Independent Discrete Gaussian Sampling for Post-Quantum Cryptography

As the development of a viable quantum computer nears, existing widely used public-key cryptosystems, such as RSA, will no longer be secure. Thus, significant effort is being invested into post-quantum cryptography (PQC). Lattice-based cryptography (LBC) is one such promising area of PQC, which offers versatile, efficient, and high performance security services. However, the vulnerabilities of these implementations against side-channel attacks (SCA) remain significantly understudied. Most, if not all, lattice-based cryptosystems require noise samples generated from a discrete Gaussian distribution, and a successful timing analysis attack can render the whole cryptosystem broken, making the discrete Gaussian sampler the most vulnerable module to SCA. This research proposes countermeasures against timing information leakage with FPGA-based designs of the CDT-based discrete Gaussian samplers with constant response time, targeting encryption and signature scheme parameters. The proposed designs are compared against the state-of-the-art and are shown to significantly outperform existing implementations. For encryption, the proposed sampler is 9x faster in comparison to the only other existing time-independent CDT sampler design. For signatures, the first time-independent CDT sampler in hardware is proposed. 

Toll-like Receptor 4 Activation Promotes Cardiac Arrhythmias By Decreasing The Transient Outward Potassium Current (ito) Through An Irf3-dependent And Myd88-independent Pathway.

Cardiac arrhythmias are one of the main causes of death worldwide. Several studies have shown that inflammation plays a key role in different cardiac diseases and Toll-like receptors (TLRs) seem to be involved in cardiac complications. In the present study, we investigated whether the activation of TLR4 induces cardiac electrical remodeling and arrhythmias, and the signaling pathway involved in these effects. Membrane potential was recorded in Wistar rat ventricle. Ca(2+) transients, as well as the L-type Ca(2+) current (ICaL) and the transient outward K(+) current (Ito), were recorded in isolated myocytes after 24 h exposure to the TLR4 agonist, lipopolysaccharide (LPS, 1 μg/ml). TLR4 stimulation in vitro promoted a cardiac electrical remodeling that leads to action potential prolongation associated with arrhythmic events, such as delayed afterdepolarization and triggered activity. After 24 h LPS incubation, Ito amplitude, as well as Kv4.3 and KChIP2 mRNA levels were reduced. The Ito decrease by LPS was prevented by inhibition of interferon regulatory factor 3 (IRF3), but not by inhibition of interleukin-1 receptor-associated kinase 4 (IRAK4) or nuclear factor kappa B (NF-κB). Extrasystolic activity was present in 25% of the cells, but apart from that, Ca(2+) transients and ICaL were not affected by LPS; however, Na(+)/Ca(2+) exchanger (NCX) activity was apparently increased. We conclude that TLR4 activation decreased Ito, which increased AP duration via a MyD88-independent, IRF3-dependent pathway. The longer action potential, associated with enhanced Ca(2+) efflux via NCX, could explain the presence of arrhythmias in the LPS group.

Comparative Metagenomic Analysis Of Coral Microbial Communities Using A Reference-independent Approach.

By comparing the SEED and Pfam functional profiles of metagenomes of two Brazilian coral species with 29 datasets that are publicly available, we were able to identify some functions, such as protein secretion systems, that are overrepresented in the metagenomes of corals and may play a role in the establishment and maintenance of bacteria-coral associations. However, only a small percentage of the reads of these metagenomes could be annotated by these reference databases, which may lead to a strong bias in the comparative studies. For this reason, we have searched for identical sequences (99% of nucleotide identity) among these metagenomes in order to perform a reference-independent comparative analysis, and we were able to identify groups of microbial communities that may be under similar selective pressures. The identification of sequences shared among the metagenomes was found to be even better for the identification of groups of communities with similar niche requirements than the traditional analysis of functional profiles. This approach is not only helpful for the investigation of similarities between microbial communities with high proportion of unknown reads, but also enables an indirect overview of gene exchange between communities.

Detection Of Explosives On The Surface Of Banknotes By Raman Hyperspectral Imaging And Independent Component Analysis.

The aim of this study was to develop a methodology using Raman hyperspectral imaging and chemometric methods for identification of pre- and post-blast explosive residues on banknote surfaces. The explosives studied were of military, commercial and propellant uses. After the acquisition of the hyperspectral imaging, independent component analysis (ICA) was applied to extract the pure spectra and the distribution of the corresponding image constituents. The performance of the methodology was evaluated by the explained variance and the lack of fit of the models, by comparing the ICA recovered spectra with the reference spectra using correlation coefficients and by the presence of rotational ambiguity in the ICA solutions. The methodology was applied to forensic samples to solve an automated teller machine explosion case. Independent component analysis proved to be a suitable method of resolving curves, achieving equivalent performance with the multivariate curve resolution with alternating least squares (MCR-ALS) method. At low concentrations, MCR-ALS presents some limitations, as it did not provide the correct solution. The detection limit of the methodology presented in this study was 50μgcm(-2).

Impact of outdoor biomass air pollution on hypertension hospital admissions

Background This study aimed to evaluate the association between the total suspended particles (TSP) generated from burning sugar cane plantations and the incidence of hospital admissions from hypertension in the city of Araraquara. Methods The study was an ecological time-series study. Total daily records of hypertension (ICD 10th I10-15) were obtained from admitted patients of all ages in a hospital in Araraquara, Sao Paulo State, Brazil, from 23 March 2003 to 27 July 2004. The daily concentration of TSP (mu g/m(3)) was obtained using a Handi-Vol sampler placed in downtown Araraquara. The local airport provided daily measures of temperature and humidity. In generalised linear Poisson regression models, the daily number of hospital admissions for hypertension was considered to be the dependent variable and the daily TSP concentration the independent variable. Results TSP presented a lagged effect on hypertension admissions, which was first observed 1 day after a TSP increase and remained almost unchanged for the following 2 days. A 10 mu g/m(3) increase in the TSP 3 day moving average lagged in 1 day led to an increase in hypertension-related hospital admissions during the harvest period (12.5%, 95% CI 5.6% to 19.9%) that was almost 30% higher than during non-harvest periods (9.0%, 95% CI 4.0% to 14.3%). Conclusions Increases in TSP concentrations were associated with hypertension-related hospital admissions. Despite the benefits of reduced air pollution in urban cities achieved by using ethanol produced from sugar cane to power automobiles, areas where the sugar cane is produced and harvested were found to have increased public health risk.

Model Independent Tests of Skyrmions and Their Holographic Cousins

We describe a new exact relation for large N(c) QCD for the long-distance behavior of baryon form factors in the chiral limit. This model-independent relation is used to test the consistency of the structure of several baryon models. All 4D semiclassical chiral soliton models satisfy the relation, as does the Pomarol-Wulzer holographic model of baryons as 5D Skyrmions. However, remarkably, we find that the holographic model treating baryons as instantons in the Sakai-Sugimoto model does not satisfy the relation.

Culture-Independent Assessment of Rhizobiales-Related Alphaproteobacteria and the Diversity of Methylobacterium in the Rhizosphere and Rhizoplane of Transgenic Eucalyptus

The rhizosphere is an ecosystem exploited by a variety of organisms involved in plant health and environmental sustainability. Abiotic factors influence microorganism-plant interactions, but the microbial community is also affected by expression of heterologous genes from host plants. In the present work, we assessed the community shifts of Alphaproteobacteria phylogenetically related to the Rhizobiales order (Rhizobiales-like community) in rhizoplane and rhizosphere soils of wild-type and transgenic eucalyptus. A greenhouse experiment was performed and the bacterial communities associated with two wild-type (WT17 and WT18) and four transgenic (TR-9, TR-15, TR-22, and TR-23) eucalyptus plant lines were evaluated. The culture-independent approach consisted of the quantification, by real-time polymerase chain reaction (PCR), of a targeted subset of Alphaproteobacteria and the assessment of its diversity using PCR-denaturing gradient gel electrophoresis (DGGE) and 16S rRNA gene clone libraries. Real-time quantification revealed a lesser density of the targeted community in TR-9 and TR-15 plants and diversity analysis by principal components analysis, based on PCR-DGGE, revealed differences between bacterial communities, not only between transgenic and nontransgenic plants, but also among wild-type plants. The comparison between clone libraries obtained from the transgenic plant TR-15 and wild-type WT17 revealed distinct bacterial communities associated with these plants. In addition, a culturable approach was used to quantify the Methylobacterium spp. in the samples where the identification of isolates, based on 16S rRNA gene sequences, showed similarities to the species Methylobacterium nodulans, Methylobacterium isbiliense, Methylobacterium variable, Methylobacterium fujisawaense, and Methylobacterium radiotolerans. Colonies classified into this genus were not isolated from the rhizosphere but brought in culture from rhizoplane samples, except for one line of the transgenic plants (TR-15). In general, the data suggested that, in most cases, shifts in bacterial communities due to cultivation of transgenic plants are similar to those observed when different wild-type cultivars are compared, although shifts directly correlated to transgenic plant cultivation may be found.

Endogenous opioids: role in prostaglandin-dependent and -independent fever

This study evaluated the participation of mu-opioid-receptor activation in body temperature (T-b) during normal and febrile conditions (including activation of heat conservation mechanisms) and in different pathways of LPS-induced fever. The intracerebroventricular treatment of male Wistar rats with the selective opioid mu-receptor-antagonist cyclic D-Phe-Cys-Try-D-Trp-Arg-Thr-Pen-Thr-NH2 (CTAP; 0.1-1.0 mu g) reduced fever induced by LPS (5.0 mu g/kg) but did not change Tb at ambient temperatures of either 20 C or 28 C. The subcutaneous, intracerebroventricular, and intrahypothalamic injection of morphine (1.0 -10.0 mg/kg, 3.0 -30.0 mu g, and 1 -100 ng, respectively) produced a dose-dependent increase in Tb. Intracerebroventricular morphine also produced a peripheral vasoconstriction. Both effects were abolished by CTAP. CTAP (1.0 mu g icv) reduced the fever induced by intracerebroventricular administration of TNF-alpha (250 ng), IL-6 (300 ng), CRF (2.5 mu g), endothelin-1 (1.0 pmol), and macrophage inflammatory protein (500 pg) and the first phase of the fever induced by PGF(2 alpha) (500.0 ng) but not the fever induced by IL-1 beta (3.12 ng) or PGE(2) (125.0 ng) or the second phase of the fever induced by PGF(2 alpha). Morphine-induced fever was not modified by the cyclooxygenase (COX) inhibitor indomethacin (2.0 mg/kg). In addition, morphine injection did not induce the expression of COX-2 in the hypothalamus, and CTAP did not modify PGE2 levels in cerebrospinal fluid or COX-2 expression in the hypothalamus after LPS injection. In conclusion, our results suggest that LPS and endogenous pyrogens (except IL-1 beta and prostaglandins) recruit the opioid system to cause a mu-receptor-mediated fever.

Relaxation evoked by extracellular Ca(2+) in rat aorta is nerve-independent and involves sarcoplasmic reticulum and L-type Ca(2+) channell

The perivascular nerve network expresses a Ca(2+) receptor that is activated by high extracellular Ca(2+) concentrations and causes vasorelaxation in resistance arteries. We have verified the influence of perivascular nerve fibers on the Ca(2+)-induced relaxation in aortic rings. To test our hypothesis, either pre-contracted aortas isolated from rats after sensory denervation with capsaicin or aortic rings acutely denervated with phenol were stimulated to relax with increasing extracellular Ca(2+) concentration. We also studied the role of the endothelium on the Ca(2+)-induced relaxation, and we verified the participation of endothelial/nonendothelial nitric oxide and cyclooxygenise-arachidonic acid metabolites. Additionally, the role of the sarcoplasmic reticulum, K(+) channels and L-type Ca(2+) channels on the Ca(2+)-induced relaxation were evaluated. We have observed that the Ca(2+)-induced relaxation is completely nerve independent, and it is potentiated by endothelial nitric oxide (NO). In endothelium-denuded aortic rings, indomethacin and AH6809 (PGF(2 alpha) receptor antagonist) enhance the relaxing response to Ca(2+). This relaxation is inhibited by thapsigargin and verapamil, while was not altered by tetraethylammonium. In conclusion, we have shown that perivascular nervous fibers do not participate in the Ca(2+)-induced relaxation, which is potentiated by endothelial NO. In endothelium-denuded preparations, indomethacin and AH6809 enhance the relaxation induced by Ca(2+). The relaxing response to Call was impaired by verapamil and thapsigargin, revealing the importance of L-type Ca(2+) channels and sarcoplasmic reticulum in this response. (c) 2008 Elsevier Inc. All rights reserved.

ATP-induced apoptosis involves a Ca(2+)-independent phospholipase A(2) and 5-lipoxygenase in macrophages

Macrophages express P2X(7) and other nucleotide (P2) receptors, and display the phenomena of extracellular ATP (ATP(e))-induced P2X(7)-dependent membrane permeabilization and cell death by apoptosis and necrosis. P2X7 receptors also cooperate with toll-like receptors (TLRs) to induce inflammasome activation and IL-1 beta secretion. We investigated signaling pathways involved in the induction of cell death by ATP, in intraperitoneal murine macrophages. Apoptosis (hypodiploid nuclei) and necrosis (LDH release) were detected 6 h after an induction period of 20 min in the presence of ATP Apoptosis was blocked by caspase 3 and caspase 9 inhibitors and by cyclosporin A. The MAPK inhibitors PD-98059, SB-203580 and SB-202190 provoked no significant effect oil apoptosis, but SB-203580 blocked LDH release. Neither apoptosis nor necrosis was inhibited when both intra- and extracellular Ca(2+) were chelated during the induction period. Mepacrine, a generic PLA(2) inhibitor and BEL, an inhibitor of Ca(2+)-independent PLA(2) (iPLA(2)) blocked apoptosis, while pBPB and AACOOPF(3). inhibitors of secretory and Ca(2+)-dependent PLA(2) respectively, had no significant effect. Cycloxygenase inhibitors had no effect on apoptosis, while the inhibitors of lipoxygenase (LOX) and leukotriene biosynthesis nordihydroguaiaretic acid (NDGA), zileuton, AA-861, and MK-886 significantly decreased apoptosis. Neither NDGA nor MK-886 blocked apoptosis of 5-LOX(-/-) macrophages. CP-105696 and MK-571, antagonists of leukotriene receptors, had no significant effect on apoptosis. None of the inhibitors of PLA(2) and LOX/leukotriene pathway had a significant inhibitory effect on LDH release. Our results indicate that a Ca(2+) -independent step involving an iPLA(2) and 5-LOX are involved in the triggering of apoptosis but not necrosis by P2X7 in macrophages. (C) 2008 Elsevier Inc. All rights reserved.

Sampling phylogenetic tree space with the generalized Gibbs sampler

The generalized Gibbs sampler (GGS) is a recently developed Markov chain Monte Carlo (MCMC) technique that enables Gibbs-like sampling of state spaces that lack a convenient representation in terms of a fixed coordinate system. This paper describes a new sampler, called the tree sampler, which uses the GGS to sample from a state space consisting of phylogenetic trees. The tree sampler is useful for a wide range of phylogenetic applications, including Bayesian, maximum likelihood, and maximum parsimony methods. A fast new algorithm to search for a maximum parsimony phylogeny is presented, using the tree sampler in the context of simulated annealing. The mathematics underlying the algorithm is explained and its time complexity is analyzed. The method is tested on two large data sets consisting of 123 sequences and 500 sequences, respectively. The new algorithm is shown to compare very favorably in terms of speed and accuracy to the program DNAPARS from the PHYLIP package.