Modelagem da irradiação direta na incidência normal em Botucatu: aprendizado de máquina, estatístico e linke

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Damage detection in beams by using artificial neural networks and dynamical parameters

In this paper is presented a multilayer perceptron neural network combined with the Nelder-Mead Simplex method to detect damage in multiple support beams. The input parameters are based on natural frequencies and modal flexibility. It was considered that only a number of modes were available and that only vertical degrees of freedom were measured. The reliability of the proposed methodology is assessed from the generation of random damages scenarios and the definition of three types of errors, which can be found during the damage identification process. Results show that the methodology can reliably determine the damage scenarios. However, its application to large beams may be limited by the high computational cost of training the neural network.

Complex network classification using partially self-avoiding deterministic walks

Complex networks have attracted increasing interest from various fields of science. It has been demonstrated that each complex network model presents specific topological structures which characterize its connectivity and dynamics. Complex network classification relies on the use of representative measurements that describe topological structures. Although there are a large number of measurements, most of them are correlated. To overcome this limitation, this paper presents a new measurement for complex network classification based on partially self-avoiding walks. We validate the measurement on a data set composed by 40000 complex networks of four well-known models. Our results indicate that the proposed measurement improves correct classification of networks compared to the traditional ones. (C) 2012 American Institute of Physics. [http://dx.doi.org/10.1063/1.4737515]

STRUCTURE AND DYNAMICS: THE TRANSITION FROM NONEQUILIBRIUM TO EQUILIBRIUM IN INTEGRATE-AND-FIRE DYNAMICS

The transient and equilibrium properties of dynamics unfolding in complex systems can depend critically on specific topological features of the underlying interconnections. In this work, we investigate such a relationship with respect to the integrate-and-fire dynamics emanating from a source node and an extended network model that allows control of the small-world feature as well as the length of the long-range connections. A systematic approach to investigate the local and global correlations between structural and dynamical features of the networks was adopted that involved extensive simulations (one and a half million cases) so as to obtain two-dimensional correlation maps. Smooth, but diverse surfaces of correlation values were obtained in all cases. Regarding the global cases, it has been verified that the onset avalanche time (but not its intensity) can be accurately predicted from the structural features within specific regions of the map (i.e. networks with specific structural properties). The analysis at local level revealed that the dynamical features before the avalanches can also be accurately predicted from structural features. This is not possible for the dynamical features after the avalanches take place. This is so because the overall topology of the network predominates over the local topology around the source at the stationary state.

Network-based high level data classification

Traditional supervised data classification considers only physical features (e. g., distance or similarity) of the input data. Here, this type of learning is called low level classification. On the other hand, the human (animal) brain performs both low and high orders of learning and it has facility in identifying patterns according to the semantic meaning of the input data. Data classification that considers not only physical attributes but also the pattern formation is, here, referred to as high level classification. In this paper, we propose a hybrid classification technique that combines both types of learning. The low level term can be implemented by any classification technique, while the high level term is realized by the extraction of features of the underlying network constructed from the input data. Thus, the former classifies the test instances by their physical features or class topologies, while the latter measures the compliance of the test instances to the pattern formation of the data. Our study shows that the proposed technique not only can realize classification according to the pattern formation, but also is able to improve the performance of traditional classification techniques. Furthermore, as the class configuration's complexity increases, such as the mixture among different classes, a larger portion of the high level term is required to get correct classification. This feature confirms that the high level classification has a special importance in complex situations of classification. Finally, we show how the proposed technique can be employed in a real-world application, where it is capable of identifying variations and distortions of handwritten digit images. As a result, it supplies an improvement in the overall pattern recognition rate.

Stochastic competitive learning in complex networks

Competitive learning is an important machine learning approach which is widely employed in artificial neural networks. In this paper, we present a rigorous definition of a new type of competitive learning scheme realized on large-scale networks. The model consists of several particles walking within the network and competing with each other to occupy as many nodes as possible, while attempting to reject intruder particles. The particle's walking rule is composed of a stochastic combination of random and preferential movements. The model has been applied to solve community detection and data clustering problems. Computer simulations reveal that the proposed technique presents high precision of community and cluster detections, as well as low computational complexity. Moreover, we have developed an efficient method for estimating the most likely number of clusters by using an evaluator index that monitors the information generated by the competition process itself. We hope this paper will provide an alternative way to the study of competitive learning.

Three-feature model to reproduce the topology of citation networks and the effects from authors' visibility on their h-index

Various factors are believed to govern the selection of references in citation networks, but a precise, quantitative determination of their importance has remained elusive. In this paper, we show that three factors can account for the referencing pattern of citation networks for two topics, namely "graphenes" and "complex networks", thus allowing one to reproduce the topological features of the networks built with papers being the nodes and the edges established by citations. The most relevant factor was content similarity, while the other two - in-degree (i.e. citation counts) and age of publication - had varying importance depending on the topic studied. This dependence indicates that additional factors could play a role. Indeed, by intuition one should expect the reputation (or visibility) of authors and/or institutions to affect the referencing pattern, and this is only indirectly considered via the in-degree that should correlate with such reputation. Because information on reputation is not readily available, we simulated its effect on artificial citation networks considering two communities with distinct fitness (visibility) parameters. One community was assumed to have twice the fitness value of the other, which amounts to a double probability for a paper being cited. While the h-index for authors in the community with larger fitness evolved with time with slightly higher values than for the control network (no fitness considered), a drastic effect was noted for the community with smaller fitness. (C) 2012 Elsevier Ltd. All rights reserved.

Mechanical Alterations in airway smooth muscle after interaction with indoor pollutant – A mathematical model.

The viscoelasticity of mammalian lung is determined by the mechanical properties and structural regulation of the airway smooth muscle (ASM). The exposure to polluted air may deteriorate these properties with harmful consequences to individual health. Formaldehyde (FA) is an important indoor pollutant found among volatile organic compounds. This pollutant permeates through the smooth muscle tissue forming covalent bonds between proteins in the extracellular matrix and intracellular protein structure changing mechanical properties of ASM and inducing asthma symptoms, such as airway hyperresponsiveness, even at low concentrations. In the experimental scenario, the mechanical effect of FA is the stiffening of the tissue, but the mechanism behind this effect is not fully understood. Thus, the aim of this study is to reproduce the mechanical behavior of the ASM, such as contraction and stretching, under FA action or not. For this, it was created a two-dimensional viscoelastic network model based on Voronoi tessellation solved using Runge-Kutta method of fourth order. The equilibrium configuration was reached when the forces in different parts of the network were equal. This model simulates the mechanical behavior of ASM through of a network of dashpots and springs. This dashpot-spring mechanical coupling mimics the composition of the actomyosin machinery of ASM through the contraction of springs to a minimum length. We hypothesized that formation of covalent bonds, due to the FA action, can be represented in the model by a simple change in the elastic constant of the springs, while the action of methacholine (MCh) reduce the equilibrium length of the spring. A sigmoid curve of tension as a function of MCh doses was obtained, showing increased tension when the muscle strip was exposed to FA. Our simulations suggest that FA, at a concentration of 0.1 ppm, can affect the elastic properties of the smooth muscle ¯bers by a factor of 120%. We also analyze the dynamic mechanical properties, observing the viscous and elastic behavior of the network. Finally, the proposed model, although simple, incorporates the phenomenology of both MCh and FA and reproduces experimental results observed with in vitro exposure of smooth muscle to FA. Thus, this new mechanical approach incorporates several well know features of the contractile system of the cells in a tissue level model. The model can also be used in different biological scales.

Mechanical alterations in airway smooth muscle after interaction with indoor pollutant - A mathematical model

The viscoelasticity of mammalian lung is determined by the mechanical properties and structural regulation of the airway smooth muscle (ASM). The exposure to polluted air may deteriorate these properties with harmful consequences to individual health. Formaldehyde (FA) is an important indoor pollutant found among volatile organic compounds. This pollutant permeates through the smooth muscle tissue forming covalent bonds between proteins in the extracellular matrix and intracellular protein structure changing mechanical properties of ASM and inducing asthma symptoms, such as airway hyperresponsiveness, even at low concentrations. In the experimental scenario, the mechanical effect of FA is the stiffening of the tissue, but the mechanism behind this effect is not fully w1derstood. Thus, the aim of this study is to reproduce the mechanical behavior of the ASM, such as contraction and stretching, under FA action or not. For this, it was created a two-dimensional viscoelastic network model based on Voronoi tessellation solved using Runge-Kutta method of fourth order. The equilibrium configuration was reached when the forces in different parts of the network were equal. This model simulates the mechanical behavior of ASM through of a network of dashpots and springs. This dashpot-spring mechanical coupling mimics the composition of the actomyosin machinery of ASM through the contraction of springs to a minimum length. We hypothesized that formation of covalent bonds, due to the FA action, can be represented in the model by a simple change in the elastic constant of the springs, while the action of methacholinc (MCh) reduce the equilibrium length of the spring. A sigmoid curve of tension as a function of MCh doses was obtained, showing increased tension when the muscle strip was exposed to FA. Our simulations suggest that FA, at a concentration of 0.1 ppm, can affect the elastic properties of the smooth muscle fibers by a factor of 120%. We also analyze the dynamic mechanical properties, observing the viscous and elastic behavior of the network. Finally, the proposed model, although simple, ir1corporates the phenomenology of both MCh and FA and reproduces experirnental results observed with ir1 vitro exposure of smooth muscle to .FA. Thus, this new mechanical approach incorporates several well know features of the contractile system of the cells ir1 a tissue level model. The model can also be used in different biological scales.

Hierarchical multi-label classification using local neural networks

Hierarchical multi-label classification is a complex classification task where the classes involved in the problem are hierarchically structured and each example may simultaneously belong to more than one class in each hierarchical level. In this paper, we extend our previous works, where we investigated a new local-based classification method that incrementally trains a multi-layer perceptron for each level of the classification hierarchy. Predictions made by a neural network in a given level are used as inputs to the neural network responsible for the prediction in the next level. We compare the proposed method with one state-of-the-art decision-tree induction method and two decision-tree induction methods, using several hierarchical multi-label classification datasets. We perform a thorough experimental analysis, showing that our method obtains competitive results to a robust global method regarding both precision and recall evaluation measures.

Variance Reduction in Analytical Chemistry : New Numerical Methods in Chemometrics and Molecular Simulation

This thesis is based on five papers addressing variance reduction in different ways. The papers have in common that they all present new numerical methods. Paper I investigates quantitative structure-retention relationships from an image processing perspective, using an artificial neural network to preprocess three-dimensional structural descriptions of the studied steroid molecules. Paper II presents a new method for computing free energies. Free energy is the quantity that determines chemical equilibria and partition coefficients. The proposed method may be used for estimating, e.g., chromatographic retention without performing experiments. Two papers (III and IV) deal with correcting deviations from bilinearity by so-called peak alignment. Bilinearity is a theoretical assumption about the distribution of instrumental data that is often violated by measured data. Deviations from bilinearity lead to increased variance, both in the data and in inferences from the data, unless invariance to the deviations is built into the model, e.g., by the use of the method proposed in paper III and extended in paper IV. Paper V addresses a generic problem in classification; namely, how to measure the goodness of different data representations, so that the best classifier may be constructed. Variance reduction is one of the pillars on which analytical chemistry rests. This thesis considers two aspects on variance reduction: before and after experiments are performed. Before experimenting, theoretical predictions of experimental outcomes may be used to direct which experiments to perform, and how to perform them (papers I and II). After experiments are performed, the variance of inferences from the measured data are affected by the method of data analysis (papers III-V).

Models for the study of neuronal activity during cognitive tasks

Assessment of brain connectivity among different brain areas during cognitive or motor tasks is a crucial problem in neuroscience today. Aim of this research study is to use neural mass models to assess the effect of various connectivity patterns in cortical EEG power spectral density (PSD), and investigate the possibility to derive connectivity circuits from EEG data. To this end, two different models have been built. In the first model an individual region of interest (ROI) has been built as the parallel arrangement of three populations, each one exhibiting a unimodal spectrum, at low, medium or high frequency. Connectivity among ROIs includes three parameters, which specify the strength of connection in the different frequency bands. Subsequent studies demonstrated that a single population can exhibit many different simultaneous rhythms, provided that some of these come from external sources (for instance, from remote regions). For this reason in the second model an individual ROI is simulated only with a single population. Both models have been validated by comparing the simulated power spectral density with that computed in some cortical regions during cognitive and motor tasks. Another research study is focused on multisensory integration of tactile and visual stimuli in the representation of the near space around the body (peripersonal space). This work describes an original neural network to simulate representation of the peripersonal space around the hands, in basal conditions and after training with a tool used to reach the far space. The model is composed of three areas for each hand, two unimodal areas (visual and tactile) connected to a third bimodal area (visual-tactile), which is activated only when a stimulus falls within the peripersonal space. Results show that the peripersonal space, which includes just a small visual space around the hand in normal conditions, becomes elongated in the direction of the tool after training, thanks to a reinforcement of synapses.

L'infiammazione intestinale nell'animale sperimentale come modello per lo sviluppo di nuovi farmaci: ruolo della via tachichininergica nelle malattie infiammatorie intestinali

Background: Several lines of evidence showed that inflammation is associated with changes in the expression of tachykinins both in human and animal models. Tachykinins, including substance P (SP), are small peptides expressed in the extrinsic primary afferent nerve fibres and enteric neurons of the gut: they exert their action through three distinct receptors, termed NK1, NK2 and NK3. SP modulates intestinal motility and enteric secretion, acting preferentially through the NK1 receptor. SP neural network and NK1 receptor expression are increased in patients with inflammatory bowel disease, and similar changes were observed in experimental models of inflammation. The 2,4 Dinitrobenzene Sulphonic Acid (DNBS) model of colitis is useful to study innate immunity, non-specific inflammation and wound healing; it has been suggested that the transmural inflammation seen in this model resembles that found in Crohn’s disease and can therefore be used to study what cells and mediators are involved in this type of inflammation. Aim: To test the possible protective effect of the NK1 receptor antagonist SSR140333 on: 1) acute model of intestinal inflammation; 2) reactivation of DNBS-induced colitis in rats. Methods: Acute colitis was induced in male SD rats by intrarectal administration of DNBS (15 mg/rat in 50% ethanol). Reactivation of colitis was induced by intrarectal injections of DNBS on day 28 (7.5 mg/rat in 35% ethanol). Animals were sacrificed on day 6 (acute colitis) and 29 (reactivation of colitis). SSR140333 (10 mg/kg) was administered orally starting from the day before the induction of colitis for 7 days (acute colitis) or seven days before the reactivation of colitis. Colonic damage was assessed by means of macroscopic and microscopic scores, myeloperoxidase activity (MPO) and TNF-α tissue levels. Enzyme immunoassay was used to measure colonic substance P levels. Statistical analysis was performed using analysis of variance (one-way or two-way, as appropriate) with the Bonferroni’s correction for multiple comparisons. Results: DNBS administration impaired body weight gain and markedly increased all inflammatory parameters (p<0.01). Treatment with SSR140333 10 mg/kg significantly counteracted the impairment in body weight gain, decreased macroscopic and histological scores and reduced colonic myeloperoxidase activity (p<0.01). Drug treatment counteracted TNF-α tissue levels and colonic SP concentrations (acute model). Similar results were obtained administering the NK1 receptor antagonist SSR140333 (3 and 10 mg/kg) for 5 days, starting the day after the induction of colitis. Intrarectal administration of DNBS four weeks after the first DNBS administration resulted in reactivation of colitis, with increases in macroscopic and histological damage scores and increase in MPO activity. Preventive treatment with SSR140333 10 mg/kg decreased macroscopic damage score, significantly reduced microscopic damage score but did not affect MPO activity. Conclusions: Treatment with SSR140333 significantly reduced intestinal damage in acute model of intestinal inflammation in rats. The NK1 receptor antagonist SSR140333 was also able to prevent relapse in experimental colitis. These results support the hypothesis of SP involvement in intestinal inflammation and indicate that NK receptor antagonists may have a therapeutic potential in inflammatory bowel disease.

Machine learning methods for prediction of disulphide bonding states of cysteine residues in proteins

The goal of this thesis work is to develop a computational method based on machine learning techniques for predicting disulfide-bonding states of cysteine residues in proteins, which is a sub-problem of a bigger and yet unsolved problem of protein structure prediction. Improvement in the prediction of disulfide bonding states of cysteine residues will help in putting a constraint in the three dimensional (3D) space of the respective protein structure, and thus will eventually help in the prediction of 3D structure of proteins. Results of this work will have direct implications in site-directed mutational studies of proteins, proteins engineering and the problem of protein folding. We have used a combination of Artificial Neural Network (ANN) and Hidden Markov Model (HMM), the so-called Hidden Neural Network (HNN) as a machine learning technique to develop our prediction method. By using different global and local features of proteins (specifically profiles, parity of cysteine residues, average cysteine conservation, correlated mutation, sub-cellular localization, and signal peptide) as inputs and considering Eukaryotes and Prokaryotes separately we have reached to a remarkable accuracy of 94% on cysteine basis for both Eukaryotic and Prokaryotic datasets, and an accuracy of 90% and 93% on protein basis for Eukaryotic dataset and Prokaryotic dataset respectively. These accuracies are best so far ever reached by any existing prediction methods, and thus our prediction method has outperformed all the previously developed approaches and therefore is more reliable. Most interesting part of this thesis work is the differences in the prediction performances of Eukaryotes and Prokaryotes at the basic level of input coding when ‘profile’ information was given as input to our prediction method. And one of the reasons for this we discover is the difference in the amino acid composition of the local environment of bonded and free cysteine residues in Eukaryotes and Prokaryotes. Eukaryotic bonded cysteine examples have a ‘symmetric-cysteine-rich’ environment, where as Prokaryotic bonded examples lack it.

Machine-learning methods for structure prediction of β-barrel membrane proteins

Different types of proteins exist with diverse functions that are essential for living organisms. An important class of proteins is represented by transmembrane proteins which are specifically designed to be inserted into biological membranes and devised to perform very important functions in the cell such as cell communication and active transport across the membrane. Transmembrane β-barrels (TMBBs) are a sub-class of membrane proteins largely under-represented in structure databases because of the extreme difficulty in experimental structure determination. For this reason, computational tools that are able to predict the structure of TMBBs are needed. In this thesis, two computational problems related to TMBBs were addressed: the detection of TMBBs in large datasets of proteins and the prediction of the topology of TMBB proteins. Firstly, a method for TMBB detection was presented based on a novel neural network framework for variable-length sequence classification. The proposed approach was validated on a non-redundant dataset of proteins. Furthermore, we carried-out genome-wide detection using the entire Escherichia coli proteome. In both experiments, the method significantly outperformed other existing state-of-the-art approaches, reaching very high PPV (92%) and MCC (0.82). Secondly, a method was also introduced for TMBB topology prediction. The proposed approach is based on grammatical modelling and probabilistic discriminative models for sequence data labeling. The method was evaluated using a newly generated dataset of 38 TMBB proteins obtained from high-resolution data in the PDB. Results have shown that the model is able to correctly predict topologies of 25 out of 38 protein chains in the dataset. When tested on previously released datasets, the performances of the proposed approach were measured as comparable or superior to the current state-of-the-art of TMBB topology prediction.