RAPD analysis of genetic variability in a multiprovenance base population of Eucalyptus grandis hill ex maiden

This study aimed to evaluate the genetic variability among individuals of a base population of Eucalyptus grandis and to build a molecular marker database for the analyzed populations. The Eucalyptus grandis base population comprised 327 individuals from Coff's Harbour, Atherton and Rio Claro. A few plants came from other sites (Belthorpe MT. Pandanus, Kenilworth, Yabbra, etc.). Since this base population had a heterogeneous composition, the groups were divided according to geographic localization (latitude and longitude), and genetic breeding level. Thus, the influence of those two factors (geographic localization and genetic breeding level) on the genetic variability detected was discussed. The RAPD technique allowed the evaluation of 70 loci. The binary matrix was used to estimate the genetic similarity among individuals using Jaccard's Coefficient. Parametric statistical tests were used to compare within-group similarity of the means. The obtained results showed that the base population had wide genetic variability and a mean genetic similarity of 0.328. Sub-group 3 (wild materials from the Atherton region) showed mean genetic similarity of 0.318. S.P.A. (from Coff's Harbour region) had a mean genetic similarity of 0.322 and was found to be very important for maintenance of variation in the base population. This can be explained since the individuals from those groups accounted for most of the base population (48.3% for it). The base population plants with genetic similarity higher than 0.60 should be phenotypically analyzed again in order to clarify the tendency of genetic variability during breeding programs.

Kernel-Based Ranking. Methods for Learning and Performance Estimation

Machine learning provides tools for automated construction of predictive models in data intensive areas of engineering and science. The family of regularized kernel methods have in the recent years become one of the mainstream approaches to machine learning, due to a number of advantages the methods share. The approach provides theoretically well-founded solutions to the problems of under- and overfitting, allows learning from structured data, and has been empirically demonstrated to yield high predictive performance on a wide range of application domains. Historically, the problems of classification and regression have gained the majority of attention in the field. In this thesis we focus on another type of learning problem, that of learning to rank. In learning to rank, the aim is from a set of past observations to learn a ranking function that can order new objects according to how well they match some underlying criterion of goodness. As an important special case of the setting, we can recover the bipartite ranking problem, corresponding to maximizing the area under the ROC curve (AUC) in binary classification. Ranking applications appear in a large variety of settings, examples encountered in this thesis include document retrieval in web search, recommender systems, information extraction and automated parsing of natural language. We consider the pairwise approach to learning to rank, where ranking models are learned by minimizing the expected probability of ranking any two randomly drawn test examples incorrectly. The development of computationally efficient kernel methods, based on this approach, has in the past proven to be challenging. Moreover, it is not clear what techniques for estimating the predictive performance of learned models are the most reliable in the ranking setting, and how the techniques can be implemented efficiently. The contributions of this thesis are as follows. First, we develop RankRLS, a computationally efficient kernel method for learning to rank, that is based on minimizing a regularized pairwise least-squares loss. In addition to training methods, we introduce a variety of algorithms for tasks such as model selection, multi-output learning, and cross-validation, based on computational shortcuts from matrix algebra. Second, we improve the fastest known training method for the linear version of the RankSVM algorithm, which is one of the most well established methods for learning to rank. Third, we study the combination of the empirical kernel map and reduced set approximation, which allows the large-scale training of kernel machines using linear solvers, and propose computationally efficient solutions to cross-validation when using the approach. Next, we explore the problem of reliable cross-validation when using AUC as a performance criterion, through an extensive simulation study. We demonstrate that the proposed leave-pair-out cross-validation approach leads to more reliable performance estimation than commonly used alternative approaches. Finally, we present a case study on applying machine learning to information extraction from biomedical literature, which combines several of the approaches considered in the thesis. The thesis is divided into two parts. Part I provides the background for the research work and summarizes the most central results, Part II consists of the five original research articles that are the main contribution of this thesis.

Logistic regression to estimate the welfare of broiler breeders in relation to environmental and behavioral variables

The increasing demand of consumer markets for the welfare of birds in poultry house has motivated many scientific researches to monitor and classify the welfare according to the production environment. Given the complexity between the birds and the environment of the aviary, the correct interpretation of the conduct becomes an important way to estimate the welfare of these birds. This study obtained multiple logistic regression models with capacity of estimating the welfare of broiler breeders in relation to the environment of the aviaries and behaviors expressed by the birds. In the experiment, were observed several behaviors expressed by breeders housed in a climatic chamber under controlled temperatures and three different ammonia concentrations from the air monitored daily. From the analysis of the data it was obtained two logistic regression models, of which the first model uses a value of ammonia concentration measured by unit and the second model uses a binary value to classify the ammonia concentration that is assigned by a person through his olfactory perception. The analysis showed that both models classified the broiler breeder's welfare successfully.

Image analysis for assessing broiler breeder behavior response to thermal environment

The research proposes a methodology for assessing broiler breeder response to changes in rearing thermal environment. The continuous video recording of a flock analyzed may offer compelling evidences of thermal comfort, as well as other indications of welfare. An algorithm for classifying specific broiler breeder behavior was developed. Videos were recorded over three boxes where 30 breeders were reared. The boxes were mounted inside an environmental chamber were ambient temperature varied from cold to hot. Digital images were processed based on the number of pixels, according to their light intensity variation and binary contrast allowing a sequence of behaviors related to welfare. The system used the default of x, y coordinates, where x represents the horizontal distance from the top left of the work area to the point P, and y is the vertical distance. The video images were observed, and a grid was developed for identifying the area the birds stayed and the time they spent at that place. The sequence was analyzed frame by frame confronting the data with specific adopted thermal neutral rearing standards. The grid mask overlapped the real bird image. The resulting image allows the visualization of clusters, as birds in flock behave in certain patterns. An algorithm indicating the breeder response to thermal environment was developed.

Finländsk transpolitik

Kirjallisuusarvostelu

Word Equations and Related Topics. Independence, Decidability and Characterizations

The three main topics of this work are independent systems and chains of word equations, parametric solutions of word equations on three unknowns, and unique decipherability in the monoid of regular languages. The most important result about independent systems is a new method giving an upper bound for their sizes in the case of three unknowns. The bound depends on the length of the shortest equation. This result has generalizations for decreasing chains and for more than three unknowns. The method also leads to shorter proofs and generalizations of some old results. Hmelevksii’s theorem states that every word equation on three unknowns has a parametric solution. We give a significantly simplified proof for this theorem. As a new result we estimate the lengths of parametric solutions and get a bound for the length of the minimal nontrivial solution and for the complexity of deciding whether such a solution exists. The unique decipherability problem asks whether given elements of some monoid form a code, that is, whether they satisfy a nontrivial equation. We give characterizations for when a collection of unary regular languages is a code. We also prove that it is undecidable whether a collection of binary regular languages is a code.

Increased oxidative stress markers may be a promising indicator of risk for primary ovarian insufficiency: a cross-sectional case control study

PURPOSE: The aim of this study was to evaluate serum levels of inducible nitric oxide synthase (INOS), myeloperoxidase (MPO), total antioxidant status (TAS), and total oxidative status (TOS) in women with primary ovarian insufficiency (POI) and to compare them with healthy fertile women. We also examined the possible risk factors associated with POI.METHODS: This cross-sectional case control study was conducted in Zekai Tahir Burak Women's Health Education and Research Hospital. The study population consisted of 44 women with POI (study group) and 36 healthy fertile women (control group). In all patients, serum levels of INOS, MPO, TAS, and TOS were determined. INOS and MPO levels were measured by enzyme-linked immunosorbent assay whereas colorimetric method was used for evaluating TAS and TOS levels. Age, body mass index (BMI), obstetric history, smoking status, family history, comorbidities, sonographic findings, complete blood count values, C-reactive protein and baseline hormone levels were also analyzed. Student's t-test or Mann-Whitney U test was used to compare continuous variables between the groups; categorical data were evaluated by using Pearson χ2 or Fisher exact test, when appropriate. Binary logistic regression method was used to identify risk factors for POI.RESULTS: We found significantly elevated levels of INOS (234.1±749.5 versus133.8±143.0; p=0.005), MPO (3,438.7±1,228.6 versus 2,481.9±1,230.1; p=0.001), and TOS (4.3±1.4 versus 3.6±1.4; p=0.02) in the sera of the study group when compared to the BMI-age matched control group. However, difference in serum levels of TAS were not significant between the 2 groups (1.7±0.2 versus 1.6±0.2; p=0.15). Logistic regression method demonstrated that BMI <25 kg/m2, nulliparity, family history of POI, smoking, and elevated serum levels of INOS, MPO, and TOS were independent risk factors for POI.CONCLUSION: We found an increase in INOS, MPO, and TOS in women with POI. These serum markers may be promising in early diagnosis of POI. Further large-scale studies are required to determine whether oxidative stress markers have a role in diagnosing POI.

Immunogenicity of an inactivated bovine herpesvirus type 5 strain defective in thymidine kinase and glycoprotein E

The immunogenicity of an inactivated, experimental vaccine based on a bovine herpesvirus type 5 strain defective in thymidine kinase and glycoprotein E (BoHV-5 gE/TKΔ) was evaluated in cattle and the results were compared with a vaccine containing the parental BoHV-5 strain (SV507/99). To formulate the vaccines, each virus (wildtype SV507/99 and BoHV-5 gE/TK∆) was multiplied in cell culture and inactivated with binary ethyleneimine (BEI). Each vaccine dose contained approximately of 10(7.5) TCID50 of inactivated virus mixed with an oil-based adjuvant (46:54). Forty calves, 6 to 9-months-old, were allocated into two groups of 20 animals each and vaccinated twice (days 0 and 22pv) by the subcutaneous route with either vaccine. Serum samples collected at day 0 and at different intervals after vaccination were tested for virus neutralizing (VN) antibodies against the parental virus and against heterologous BoHV-5 and BoHV-1 isolates. The VN assays demonstrated seroconversion to the respective homologous viruses in all vaccinated animals after the second vaccine dose (mean titers of 17.5 for the wildtype vaccine; 24.1 for the recombinant virus). All animals remained reagents up to day 116 pv, yet showing a gradual reduction in VN titers. Animals from both vaccine groups reacted in similar VN titers to different BoHV-1 and BoHV-5 isolates, yet the magnitude of serological response of both groups was higher against BoHV-5 field isolates. Calves vaccinated with the recombinant virus did not develop antibodies to gE as verified by negative results in a gE-specific ELISA, what would allow serological differentiation from naturally infected animals. Taken together, these results indicate that inactivated antigens of BoHV-5 gE/TK recombinant virus induced an adequate serological response against BoHV-5 and BoHV-1 and thus can be used as an alternative, differential vaccine candidate.

Evaluation of follicular lymphoid depletion in the Bursa of Fabricius: an alternative methodology using digital image analysis and artificial neural networks

Fifty Bursa of Fabricius (BF) were examined by conventional optical microscopy and digital images were acquired and processed using Matlab® 6.5 software. The Artificial Neuronal Network (ANN) was generated using Neuroshell® Classifier software and the optical and digital data were compared. The ANN was able to make a comparable classification of digital and optical scores. The use of ANN was able to classify correctly the majority of the follicles, reaching sensibility and specificity of 89% and 96%, respectively. When the follicles were scored and grouped in a binary fashion the sensibility increased to 90% and obtained the maximum value for the specificity of 92%. These results demonstrate that the use of digital image analysis and ANN is a useful tool for the pathological classification of the BF lymphoid depletion. In addition it provides objective results that allow measuring the dimension of the error in the diagnosis and classification therefore making comparison between databases feasible.

Constitutive laws for strong geometric non-linearity

This paper is divided into two different parts. The first one provides a brief introduction to the fractal geometry with some simple illustrations in fluid mechanics. We thought it would be helpful to introduce the reader into this relatively new approach to mechanics that has not been sufficiently explored by engineers yet. Although in fluid mechanics, mainly in problems of percolation and binary flows, the use of fractals has gained some attention, the same is not true for solid mechanics, from the best of our knowledge. The second part deals with the mechanical behavior of thin wires subjected to very large deformations. It is shown that starting to a plausible conjecture it is possible to find global constitutive equations correlating geometrical end energy variables with the fractal dimension of the solid subjected to large deformations. It is pointed out the need to complement the present proposal with experimental work.

Switchable Lanthanide Luminescence for Detection of Biomolecules

Binary probes are oligonucleotide probe pairs that hybridize adjacently to a complementary target nucleic acid. In order to detect this hybridization, the two probes can be modified with, for example, fluorescent molecules, chemically reactive groups or nucleic acid enzymes. The benefit of this kind of binary probe based approach is that the hybridization elicits a detectable signal which is distinguishable from background noise even though unbound probes are not removed by washing before measurement. In addition, the requirement of two simultaneous binding events increases specificity. Similarly to binary oligonucleotide probes, also certain enzymes and fluorescent proteins can be divided into two parts and used in separation-free assays. Split enzyme and fluorescent protein reporters have practical applications among others as tools to investigate protein-protein interactions within living cells. In this study, a novel label technology, switchable lanthanide luminescence, was introduced and used successfully in model assays for nucleic acid and protein detection. This label technology is based on a luminescent lanthanide chelate divided into two inherently non-luminescent moieties, an ion carrier chelate and a light harvesting antenna ligand. These form a highly luminescent complex when brought into close proximity; i.e., the label moieties switch from a dark state to a luminescent state. This kind of mixed lanthanide complex has the same beneficial photophysical properties as the more typical lanthanide chelates and cryptates - sharp emission peaks, long emission lifetime enabling time-resolved measurement, and large Stokes’ shift, which minimize the background signal. Furthermore, the switchable lanthanide luminescence technique enables a homogeneous assay set-up. Here, switchable lanthanide luminescence label technology was first applied to sensitive, homogeneous, single-target nucleic acid and protein assays with picomolar detection limits and high signal to background ratios. Thereafter, a homogeneous four-plex nucleic acid array-based assay was developed. Finally, the label technology was shown to be effective in discrimination of single nucleotide mismatched targets from fully matched targets and the luminescent complex formation was analyzed more thoroughly. In conclusion, this study demonstrates that the switchable lanthanide luminescencebased label technology can be used in various homogeneous bioanalytical assays.

Combinatorics on Words. New Aspects on Avoidability, Defect Effect, Equations and Palindromes

In this thesis we examine four well-known and traditional concepts of combinatorics on words. However the contexts in which these topics are treated are not the traditional ones. More precisely, the question of avoidability is asked, for example, in terms of k-abelian squares. Two words are said to be k-abelian equivalent if they have the same number of occurrences of each factor up to length k. Consequently, k-abelian equivalence can be seen as a sharpening of abelian equivalence. This fairly new concept is discussed broader than the other topics of this thesis. The second main subject concerns the defect property. The defect theorem is a well-known result for words. We will analyze the property, for example, among the sets of 2-dimensional words, i.e., polyominoes composed of labelled unit squares. From the defect effect we move to equations. We will use a special way to define a product operation for words and then solve a few basic equations over constructed partial semigroup. We will also consider the satisfiability question and the compactness property with respect to this kind of equations. The final topic of the thesis deals with palindromes. Some finite words, including all binary words, are uniquely determined up to word isomorphism by the position and length of some of its palindromic factors. The famous Thue-Morse word has the property that for each positive integer n, there exists a factor which cannot be generated by fewer than n palindromes. We prove that in general, every non ultimately periodic word contains a factor which cannot be generated by fewer than 3 palindromes, and we obtain a classification of those binary words each of whose factors are generated by at most 3 palindromes. Surprisingly these words are related to another much studied set of words, Sturmian words.

Local Features in Image and Video Processing - Object Class Matching and Video Shot Detection

The usage of digital content, such as video clips and images, has increased dramatically during the last decade. Local image features have been applied increasingly in various image and video retrieval applications. This thesis evaluates local features and applies them to image and video processing tasks. The results of the study show that 1) the performance of different local feature detector and descriptor methods vary significantly in object class matching, 2) local features can be applied in image alignment with superior results against the state-of-the-art, 3) the local feature based shot boundary detection method produces promising results, and 4) the local feature based hierarchical video summarization method shows promising new new research direction. In conclusion, this thesis presents the local features as a powerful tool in many applications and the imminent future work should concentrate on improving the quality of the local features.

Infestation of weed species in monocrop coffee or intercropped with banana, under agroecological system

We aimed with this study to compare weed infestation in coffee under two different cropping managements: conventional coffee grown alone, or intercropped with banana plantation in a year-round basis (late spring, late summer, late fall and late winter). The experiment was installed in 2009 under field conditions at the Escola Municipal Rural Benedita Figueiró de Oliveira, in the city of Ivinhema in the state of Mato Grosso do Sul, Brazil. Assessments of weed occurrence were made three years after employment, on both cropping systems, and density, frequency, dominance and the importance value for each plant species in each system and season were quantified. Plant diversity within each system was estimated by Simpson and Shannon-Weiner indexes. Similarity between cropping systems were also assessed by the binary asymmetric similarity coefficient of Jaccard. Absolute infestation and spontaneous species differed between the two cropping systems in all seasons. Overall species diversity is higher in the monocrop compared with the intercrop, and it is associated in this study with the higher incidence of troublesome species. Areas were similar in terms of weed composition only in the Fall. Shading provided by the banana trees shows to be an efficient culture management aiming to suppress weeds in agro-ecological planting systems.