BME-based uncertainty assessment of the Chernobyl fallout

The vast territories that have been radioactively contaminated during the 1986 Chernobyl accident provide a substantial data set of radioactive monitoring data, which can be used for the verification and testing of the different spatial estimation (prediction) methods involved in risk assessment studies. Using the Chernobyl data set for such a purpose is motivated by its heterogeneous spatial structure (the data are characterized by large-scale correlations, short-scale variability, spotty features, etc.). The present work is concerned with the application of the Bayesian Maximum Entropy (BME) method to estimate the extent and the magnitude of the radioactive soil contamination by 137Cs due to the Chernobyl fallout. The powerful BME method allows rigorous incorporation of a wide variety of knowledge bases into the spatial estimation procedure leading to informative contamination maps. Exact measurements (?hard? data) are combined with secondary information on local uncertainties (treated as ?soft? data) to generate science-based uncertainty assessment of soil contamination estimates at unsampled locations. BME describes uncertainty in terms of the posterior probability distributions generated across space, whereas no assumption about the underlying distribution is made and non-linear estimators are automatically incorporated. Traditional estimation variances based on the assumption of an underlying Gaussian distribution (analogous, e.g., to the kriging variance) can be derived as a special case of the BME uncertainty analysis. The BME estimates obtained using hard and soft data are compared with the BME estimates obtained using only hard data. The comparison involves both the accuracy of the estimation maps using the exact data and the assessment of the associated uncertainty using repeated measurements. Furthermore, a comparison of the spatial estimation accuracy obtained by the two methods was carried out using a validation data set of hard data. Finally, a separate uncertainty analysis was conducted that evaluated the ability of the posterior probabilities to reproduce the distribution of the raw repeated measurements available in certain populated sites. The analysis provides an illustration of the improvement in mapping accuracy obtained by adding soft data to the existing hard data and, in general, demonstrates that the BME method performs well both in terms of estimation accuracy as well as in terms estimation error assessment, which are both useful features for the Chernobyl fallout study.

Caracterização das Comunidades Subtidais de Macroinvertebrados Bentónicos de Substrato Rochoso da Ilha de São Vicente, Cabo Verde

The Cape Verde Archipelago location and its biogeographical features are of special interest for Marine Ecology. However, there’s a lack of knowledge regarding the composition of the coastal ecosystems in this region, especially about benthic macroinvertebrates subtidal communities. Between August and October of 2007, eight locations around the island of São Vicente were sampled. Within each of those spots, fragments of substratum were collected and throughout the processing of the collected data, a total of 4032 individuals were counted, which belong to 81 different species. Shannon’s Entropy and Gini-Simpson’s diversity index were calculated, as the real number of species each one represented. By comparing the results, differences between sampling stations and between indices within the same sampling station were found. With the purpose of clustering the sampled locations according to the number of collected organisms by species, a dendrogram was elaborated and a principal component analysis was carried out. The considered sampling stations didn’t reveal significant differences according to the composition of their benthic macroinvertebrates subtidal communities in terms of great taxonomic groups or functional groups. It’s assumed that they differ only by minute traits.

Adult sex ratio effects on male survivorship of Drosophila melanogaster Meigen (Diptera, Drosophilidae)

The behavioral biology has a central role in evolutionary biology mainly because the antagonistic relations that occur in the sexual reproduction. One involves the effect of reproduction on the future life expectation. In this scenario, changes in male operational sex ratio could lead to an increase in mortality due to costs associated with excessive courtship and mating displays. Thus, this work experimentally altered the male sex ratio of Drosophila melanogaster Meigen, 1830, to determine its impact on mortality. The results indicated that mortality increases as the sex ratio changes, including modifications in the survivorship curve type and in the curve concavity, measured by entropy.

Worst-case bounds for the logarithmic loss of predictors

We investigate on-line prediction of individual sequences. Given a class of predictors, the goal is to predict as well as the best predictor in the class, where the loss is measured by the self information (logarithmic) loss function. The excess loss (regret) is closely related to the redundancy of the associated lossless universal code. Using Shtarkov's theorem and tools from empirical process theory, we prove a general upper bound on the best possible (minimax) regret. The bound depends on certain metric properties of the class of predictors. We apply the bound to both parametric and nonparametric classes ofpredictors. Finally, we point out a suboptimal behavior of the popular Bayesian weighted average algorithm.

Model selection and error estimation

We study model selection strategies based on penalized empirical loss minimization. We point out a tight relationship between error estimation and data-based complexity penalization: any good error estimate may be converted into a data-based penalty function and the performance of the estimate is governed by the quality of the error estimate. We consider several penalty functions, involving error estimates on independent test data, empirical {\sc vc} dimension, empirical {\sc vc} entropy, andmargin-based quantities. We also consider the maximal difference between the error on the first half of the training data and the second half, and the expected maximal discrepancy, a closely related capacity estimate that can be calculated by Monte Carlo integration. Maximal discrepancy penalty functions are appealing for pattern classification problems, since their computation is equivalent to empirical risk minimization over the training data with some labels flipped.

Characterization of a major V3 duplication in the NS5A region of genotype 1b HCV by quasispecies analysis in a French multicenter study

Background and Aims: The NS5A protein of the HCV is known tobe involved in viral replication and assembly and probably in theresistance to Interferon based-therapy. Previous studies identifiedinsertions or deletions from 1 to 12 nucleotides in several genomicregions. In a multicenter study (17 French and 1 Swiss laboratoriesof virology), we identified for the first time a 31 amino acidsinsertion leading to a duplication of the V3 domain in the NS5Aregion with a high prevalence. Quasispecies of each strain withduplication were characterized and the inserted V3 domain wasidentified.Methods: Between 2006 and 2008, 1067 patients chronicallyinfected with a 1b HCV were consecutively included in the study.We first amplified the V3 region by RT-PCR to detect duplication(919 samples successfully amplified). The entire NS5A region wasthen amplified, cloned and sequenced in strains bearing theduplication. V3 sequences (called R1 and R2) from each clonewere analyzed with BioEdit and compared to a V3 consensussequence (C) built from the Database Los Alamos Hepatitis C.Entropy was determined at each position.Results: V3 duplications were identified in 25 patients representinga prevalence of 2.72%. We sequenced 2043 clones from which776 had a complete coding NS5A sequence (corresponding toa mean of 30 clones per patient). At the intra-individual level,6 to 17 variants were identified per V3 region, with a maximum of3 different amino acids. At the inter-individual level, a differenceof 7 and 2 amino acids was observed between C and R1 and R2sequences, respectively. Moreover few positions presented entropyhigher than 1 (4 for the R1, 2 for the R2 and 2 for the C). Among allthe sequenced clones, more than 60% were defective virus (partialfragment of NS5A or stop codon).Conclusions: We identified a duplication of the V3 domain ingenotype 1b HCV with a high prevalence. The R2 domain, which wasthe most similar to the C region, might probably be the "original"domain, whereas R1 should be the inserted domain. Phylogeneticanalyses are under process to confirm this hypothesis.

Active learning methods for remote sensing image classification

In this paper, we propose two active learning algorithms for semiautomatic definition of training samples in remote sensing image classification. Based on predefined heuristics, the classifier ranks the unlabeled pixels and automatically chooses those that are considered the most valuable for its improvement. Once the pixels have been selected, the analyst labels them manually and the process is iterated. Starting with a small and nonoptimal training set, the model itself builds the optimal set of samples which minimizes the classification error. We have applied the proposed algorithms to a variety of remote sensing data, including very high resolution and hyperspectral images, using support vector machines. Experimental results confirm the consistency of the methods. The required number of training samples can be reduced to 10% using the methods proposed, reaching the same level of accuracy as larger data sets. A comparison with a state-of-the-art active learning method, margin sampling, is provided, highlighting advantages of the methods proposed. The effect of spatial resolution and separability of the classes on the quality of the selection of pixels is also discussed.

Automated analysis of background EEG and reactivity during therapeutic hypothermia in comatose patients after cardiac arrest.

Visual analysis of electroencephalography (EEG) background and reactivity during therapeutic hypothermia provides important outcome information, but is time-consuming and not always consistent between reviewers. Automated EEG analysis may help quantify the brain damage. Forty-six comatose patients in therapeutic hypothermia, after cardiac arrest, were included in the study. EEG background was quantified with burst-suppression ratio (BSR) and approximate entropy, both used to monitor anesthesia. Reactivity was detected through change in the power spectrum of signal before and after stimulation. Automatic results obtained almost perfect agreement (discontinuity) to substantial agreement (background reactivity) with a visual score from EEG-certified neurologists. Burst-suppression ratio was more suited to distinguish continuous EEG background from burst-suppression than approximate entropy in this specific population. Automatic EEG background and reactivity measures were significantly related to good and poor outcome. We conclude that quantitative EEG measurements can provide promising information regarding current state of the patient and clinical outcome, but further work is needed before routine application in a clinical setting.

Brownian dynamics simulation of DNA condensation.

DNA condensation observed in vitro with the addition of polyvalent counterions is due to intermolecular attractive forces. We introduce a quantitative model of these forces in a Brownian dynamics simulation in addition to a standard mean-field Poisson-Boltzmann repulsion. The comparison of a theoretical value of the effective diameter calculated from the second virial coefficient in cylindrical geometry with some experimental results allows a quantitative evaluation of the one-parameter attractive potential. We show afterward that with a sufficient concentration of divalent salt (typically approximately 20 mM MgCl(2)), supercoiled DNA adopts a collapsed form where opposing segments of interwound regions present zones of lateral contact. However, under the same conditions the same plasmid without torsional stress does not collapse. The condensed molecules present coexisting open and collapsed plectonemic regions. Furthermore, simulations show that circular DNA in 50% methanol solutions with 20 mM MgCl(2) aggregates without the requirement of torsional energy. This confirms known experimental results. Finally, a simulated DNA molecule confined in a box of variable size also presents some local collapsed zones in 20 mM MgCl(2) above a critical concentration of the DNA. Conformational entropy reduction obtained either by supercoiling or by confinement seems thus to play a crucial role in all forms of condensation of DNA.

Extremal climatic states simulated by a 2-dimensional model Part I: Sensitivity of the model and present state

The criterion, based on the thermodynamics theory, that the climatic system tends to extremizesome function has suggested several studies. In particular, special attention has been devoted to the possibility that the climate reaches an extremal rate of planetary entropy production.Due to both radiative and material effects contribute to total planetary entropy production,climatic simulations obtained at the extremal rates of total, radiative or material entropy production appear to be of interest in order to elucidate which of the three extremal assumptions behaves more similar to current data. In the present paper, these results have been obtainedby applying a 2-dimensional (2-Dim) horizontal energy balance box-model, with a few independent variables (surface temperature, cloud-cover and material heat fluxes). In addition, climatic simulations for current conditions by assuming a fixed cloud-cover have been obtained. Finally,sensitivity analyses for both variable and fixed cloud models have been carried out

Extremal climatic states simulated by a 2-dimensional model Part II: Different climatic scenarios

Different climatic simulations have been obtained by using a 2-Dim horizontal energy balancemodel (EBM), which has been constrained to satisfy several extremal principles on dissipationand convection. Moreover, 2 different versions of the model with fixed and variable cloud-coverhave been used. The assumption of an extremal type of behaviour for the climatic system canacquire additional support depending on the similarities found with measured data for pastconditions as well as with usual projections for possible future scenarios

Bayesian image reconstruction with space-variant noise suppression

In this paper we present a Bayesian image reconstruction algorithm with entropy prior (FMAPE) that uses a space-variant hyperparameter. The spatial variation of the hyperparameter allows different degrees of resolution in areas of different statistical characteristics, thus avoiding the large residuals resulting from algorithms that use a constant hyperparameter. In the first implementation of the algorithm, we begin by segmenting a Maximum Likelihood Estimator (MLE) reconstruction. The segmentation method is based on using a wavelet decomposition and a self-organizing neural network. The result is a predetermined number of extended regions plus a small region for each star or bright object. To assign a different value of the hyperparameter to each extended region and star, we use either feasibility tests or cross-validation methods. Once the set of hyperparameters is obtained, we carried out the final Bayesian reconstruction, leading to a reconstruction with decreased bias and excellent visual characteristics. The method has been applied to data from the non-refurbished Hubble Space Telescope. The method can be also applied to ground-based images.

Simulations in evolution. II. Relative fitness and the propagation of mutants.

In Neo-Darwinism, variation and natural selection are the two evolutionary mechanisms which propel biological evolution. Our previous article presented a histogram model [1] consisting in populations of individuals whose number changed under the influence of variation and/or fitness, the total population remaining constant. Individuals are classified into bins, and the content of each bin is calculated generation after generation by an Excel spreadsheet. Here, we apply the histogram model to a stable population with fitness F(1)=1.00 in which one or two fitter mutants emerge. In a first scenario, a single mutant emerged in the population whose fitness was greater than 1.00. The simulations ended when the original population was reduced to a single individual. The histogram model was validated by excellent agreement between its predictions and those of a classical continuous function (Eqn. 1) which predicts the number of generations needed for a favorable mutation to spread throughout a population. But in contrast to Eqn. 1, our histogram model is adaptable to more complex scenarios, as demonstrated here. In the second and third scenarios, the original population was present at time zero together with two mutants which differed from the original population by two higher and distinct fitness values. In the fourth scenario, the large original population was present at time zero together with one fitter mutant. After a number of generations, when the mutant offspring had multiplied, a second mutant was introduced whose fitness was even greater. The histogram model also allows Shannon entropy (SE) to be monitored continuously as the information content of the total population decreases or increases. The results of these simulations illustrate, in a graphically didactic manner, the influence of natural selection, operating through relative fitness, in the emergence and dominance of a fitter mutant.

Diffusionless first-order phase transitions in systems with frozen configurational degrees of freedom

We consider systems that can be described in terms of two kinds of degree of freedom. The corresponding ordering modes may, under certain conditions, be coupled to each other. We may thus assume that the primary ordering mode gives rise to a diffusionless first-order phase transition. The change of its thermodynamic properties as a function of the secondary-ordering-mode state is then analyzed. Two specific examples are discussed. First, we study a three-state Potts model in a binary system. Using mean-field techniques, we obtain the phase diagram and different properties of the system as a function of the distribution of atoms on the different lattice sites. In the second case, the properties of a displacive structural phase transition of martensitic type in a binary alloy are studied as a function of atomic order. Because of the directional character of the martensitic-transition mechanism, we find only a very weak dependence of the entropy on atomic order. Experimental results are found to be in quite good agreement with theoretical predictions.

Influence of configurational atomic order on the relative stability of bcc and close-packed structures in Cu-based alloys

We present a study of the influence of atomic order on the relative stability of the bcc and the 18R martensitic structures in a Cu2.96Al0.92Be0.12 crystal. Calorimetric measurements have shown that disorder increases the stability of the 18R phase, contrary to what happens in Cu-Zn-Al alloys for which it is the bcc phase that is stabilized by disordering the system. This different behavior has been explained in terms of a model recently reported. We have also proved that the entropy change at the martensitic transition is independent of the state of atomic order of the crystal, as predicted theoretically. Our results suggest that differences in the vibrational spectrum of the crystal due to different states of atomic order must be equal in the bcc and in the close-packed phases.