Automatic learning for the classification of chemical reactions and in statistical thermodynamics

This Thesis describes the application of automatic learning methods for a) the classification of organic and metabolic reactions, and b) the mapping of Potential Energy Surfaces(PES). The classification of reactions was approached with two distinct methodologies: a representation of chemical reactions based on NMR data, and a representation of chemical reactions from the reaction equation based on the physico-chemical and topological features of chemical bonds. NMR-based classification of photochemical and enzymatic reactions. Photochemical and metabolic reactions were classified by Kohonen Self-Organizing Maps (Kohonen SOMs) and Random Forests (RFs) taking as input the difference between the 1H NMR spectra of the products and the reactants. The development of such a representation can be applied in automatic analysis of changes in the 1H NMR spectrum of a mixture and their interpretation in terms of the chemical reactions taking place. Examples of possible applications are the monitoring of reaction processes, evaluation of the stability of chemicals, or even the interpretation of metabonomic data. A Kohonen SOM trained with a data set of metabolic reactions catalysed by transferases was able to correctly classify 75% of an independent test set in terms of the EC number subclass. Random Forests improved the correct predictions to 79%. With photochemical reactions classified into 7 groups, an independent test set was classified with 86-93% accuracy. The data set of photochemical reactions was also used to simulate mixtures with two reactions occurring simultaneously. Kohonen SOMs and Feed-Forward Neural Networks (FFNNs) were trained to classify the reactions occurring in a mixture based on the 1H NMR spectra of the products and reactants. Kohonen SOMs allowed the correct assignment of 53-63% of the mixtures (in a test set). Counter-Propagation Neural Networks (CPNNs) gave origin to similar results. The use of supervised learning techniques allowed an improvement in the results. They were improved to 77% of correct assignments when an ensemble of ten FFNNs were used and to 80% when Random Forests were used. This study was performed with NMR data simulated from the molecular structure by the SPINUS program. In the design of one test set, simulated data was combined with experimental data. The results support the proposal of linking databases of chemical reactions to experimental or simulated NMR data for automatic classification of reactions and mixtures of reactions. Genome-scale classification of enzymatic reactions from their reaction equation. The MOLMAP descriptor relies on a Kohonen SOM that defines types of bonds on the basis of their physico-chemical and topological properties. The MOLMAP descriptor of a molecule represents the types of bonds available in that molecule. The MOLMAP descriptor of a reaction is defined as the difference between the MOLMAPs of the products and the reactants, and numerically encodes the pattern of bonds that are broken, changed, and made during a chemical reaction. The automatic perception of chemical similarities between metabolic reactions is required for a variety of applications ranging from the computer validation of classification systems, genome-scale reconstruction (or comparison) of metabolic pathways, to the classification of enzymatic mechanisms. Catalytic functions of proteins are generally described by the EC numbers that are simultaneously employed as identifiers of reactions, enzymes, and enzyme genes, thus linking metabolic and genomic information. Different methods should be available to automatically compare metabolic reactions and for the automatic assignment of EC numbers to reactions still not officially classified. In this study, the genome-scale data set of enzymatic reactions available in the KEGG database was encoded by the MOLMAP descriptors, and was submitted to Kohonen SOMs to compare the resulting map with the official EC number classification, to explore the possibility of predicting EC numbers from the reaction equation, and to assess the internal consistency of the EC classification at the class level. A general agreement with the EC classification was observed, i.e. a relationship between the similarity of MOLMAPs and the similarity of EC numbers. At the same time, MOLMAPs were able to discriminate between EC sub-subclasses. EC numbers could be assigned at the class, subclass, and sub-subclass levels with accuracies up to 92%, 80%, and 70% for independent test sets. The correspondence between chemical similarity of metabolic reactions and their MOLMAP descriptors was applied to the identification of a number of reactions mapped into the same neuron but belonging to different EC classes, which demonstrated the ability of the MOLMAP/SOM approach to verify the internal consistency of classifications in databases of metabolic reactions. RFs were also used to assign the four levels of the EC hierarchy from the reaction equation. EC numbers were correctly assigned in 95%, 90%, 85% and 86% of the cases (for independent test sets) at the class, subclass, sub-subclass and full EC number level,respectively. Experiments for the classification of reactions from the main reactants and products were performed with RFs - EC numbers were assigned at the class, subclass and sub-subclass level with accuracies of 78%, 74% and 63%, respectively. In the course of the experiments with metabolic reactions we suggested that the MOLMAP / SOM concept could be extended to the representation of other levels of metabolic information such as metabolic pathways. Following the MOLMAP idea, the pattern of neurons activated by the reactions of a metabolic pathway is a representation of the reactions involved in that pathway - a descriptor of the metabolic pathway. This reasoning enabled the comparison of different pathways, the automatic classification of pathways, and a classification of organisms based on their biochemical machinery. The three levels of classification (from bonds to metabolic pathways) allowed to map and perceive chemical similarities between metabolic pathways even for pathways of different types of metabolism and pathways that do not share similarities in terms of EC numbers. Mapping of PES by neural networks (NNs). In a first series of experiments, ensembles of Feed-Forward NNs (EnsFFNNs) and Associative Neural Networks (ASNNs) were trained to reproduce PES represented by the Lennard-Jones (LJ) analytical potential function. The accuracy of the method was assessed by comparing the results of molecular dynamics simulations (thermal, structural, and dynamic properties) obtained from the NNs-PES and from the LJ function. The results indicated that for LJ-type potentials, NNs can be trained to generate accurate PES to be used in molecular simulations. EnsFFNNs and ASNNs gave better results than single FFNNs. A remarkable ability of the NNs models to interpolate between distant curves and accurately reproduce potentials to be used in molecular simulations is shown. The purpose of the first study was to systematically analyse the accuracy of different NNs. Our main motivation, however, is reflected in the next study: the mapping of multidimensional PES by NNs to simulate, by Molecular Dynamics or Monte Carlo, the adsorption and self-assembly of solvated organic molecules on noble-metal electrodes. Indeed, for such complex and heterogeneous systems the development of suitable analytical functions that fit quantum mechanical interaction energies is a non-trivial or even impossible task. The data consisted of energy values, from Density Functional Theory (DFT) calculations, at different distances, for several molecular orientations and three electrode adsorption sites. The results indicate that NNs require a data set large enough to cover well the diversity of possible interaction sites, distances, and orientations. NNs trained with such data sets can perform equally well or even better than analytical functions. Therefore, they can be used in molecular simulations, particularly for the ethanol/Au (111) interface which is the case studied in the present Thesis. Once properly trained, the networks are able to produce, as output, any required number of energy points for accurate interpolations.

Valorization of traditional portuguese apples and cherries: biochemical characterization and development of functional ingredients

Dissertation presented to obtain a Ph.D degree in Engineering and Technology Sciences, Biotechnology at the Instituto de Tecnologia Química e Biológica, Universidade Nova de Lisboa

Proteomic analysis of a neurodegeneration cell model, treated with plant extracts with potential neuroprotective activity

Nowadays, a significant increase in chronic diseases is observed. Epidemiological studies showed a consistent relationship between the consumption of fruits and vegetables and a reduced risk of certain chronic diseases, namely neurodegenerative disorders. One factor common to these diseases is oxidative stress, which is highly related with proteins, lipids, carbohydrates and nucleic acids damage, leading to cellular dysfunction. Polyphenols, highly abundant in berries and associated products, were described as having antioxidant properties, with beneficial effect in these pathologies. The aims of this study were to evaluate by proteomic analyses the effect of oxidative insult in a neuroblastoma cell line (SK-N-MC) and understand the mechanisms involved in the neuroprotective effects of digested extracts from commercial and wild blackberry (R. vagabundus Samp.). The analysis of the total proteome by two-dimensional electrophoresis revealed that oxidative stress in SK-N-MC cells resulted in altered expression of 12 protein spots from a total of 318. Regarding some redox proteomics alterations, particularly proteins carbonylation and glutathionylation, protein carbonyl alterations during stress suggest that cells produce an early and late response; on the other hand, no glutathionylated polypeptides were detected. Relatively to the incubation of SK-N-MC cells with digested berry extracts, commercial blackberry promotes more changes in protein pattern of these cells than R. vagabundus. From 9 statistically different protein spots of cells incubated with commercial blackberry, only β-tubulin and GRP 78 were until now identified by mass spectrometry. Further studies involving the selection of sub proteomes will be necessary to have a better understanding of the mechanisms underlying the neuroprotective effects of berries.

Alimentação e estado nutricional de grupos da população do concelho de Lisboa

RESUMO - Objectivo: descrever o estado nutricional e o padrão alimentar de grupos da população de Lisboa. Delineamento, locais e participantes: estudo transversal do estado nutricional e do padrão alimentar de homens com 38 anos e mulheres e homens com 50-65 anos que vivem na freguesia do Lumiar, do concelho de Lisboa; estudo transversal dos hábitos alimentares e do estado nutricional de adolescentes com 12 a 19 anos estudantes de escolas secundárias de Lisboa; estudo transversal da excreção de sódio em urina de 24 horas de rapazes de 7 a 8 anos de escolas primárias do Lumiar. Métodos: o estado nutricional foi avaliado com metodologias padronizadas tal como propostas pela OMS. A ingestão alimentar foi medida pelo método das 24 horas anteriores no primeiro estudo e pelo método da história alimentar no segundo estudo. Resultados: nas crianças, a excreção urinária de sódio foi uma das mais altas da Europa. Nos adolescentes, a prevalência de excesso de peso/obesidade foi de 19% nos rapazes e 16% nas raparigas. 16% dos adolescentes não tomam pequeno-almoço, mas mais de 30% comem bolos a meio da manhã. A proporção de adolescentes com uma ingestão diária de certos alimentos é de 44% para guloseimas, 43% para bolos, 29% para refrigerantes e 16% para chocolates e gelados. Por outro lado, a frequência de ingestão diária de fruta e produtos hortícolas é baixa, principalmente nos adolescentes de menor nível sócio-económico. Nos adultos, 57% dos homens com 38 anos e 80% e 74% dos homens e mulheres com 50-65 anos, respectivamente, têm excesso de peso ou obesidade. A hipertensão arterial foi detectada em mais de 20% dos homens jovens e a sua prevalência aumenta com a idade. Foi observado um colesterol sérico superior a 200 mg/dL em perto de metade dos adultos e maior do que 240 mg/dL em 28% dos homens com 38 anos, em 24% dos homens com 50-65 anos e 44% das mulheres do mesmo grupo etário. 20% dos adultos jovens têm colesterol HDL inferior a 35 mg/dL, mais do que um terço são fumadores e a ingestão de álcool representa 13% da sua ingestão calórica total. Conclusões: foram observados estilos de vida não saudáveis, nomeadamente padrões alimentares não equilibrados, uma elevada prevalência de obesidade, hipertensão e dislipidemias, e, por isso, uma percentagem relativamente elevada da população de Lisboa tem um risco elevado de doenças cárdio-vasculares e outras doenças crónicas, bem como de morte prematura.

An overview of the higher level classification of Pucciniomycotina based on combined analyses of nuclear large and small subunit rDNA sequences

Mycologia, Vol. 98, nº6

Unsupervised classification of remote sensing images combining Self Organizing Maps and segmentation techniques

Dissertation submitted in partial fulfillment of the requirements for the Degree of Master of Science in Geospatial Technologies.

Creation and extension of ontologies for describing communications in the context of organizations

Thesis submitted to Faculdade de Ciências e Tecnologia of the Universidade Nova de Lisboa, in partial fulfillment of the requirements for the degree of Master in Computer Science

Evaluation of Opuntia spp. bioactive products as promising natural chemotherapeutical agents- an in vitro approach

Dissertation to obtain a Master Degree in Biotechnology

Evaluation of chemotherapeutic potential of natural extracts using 3D models of colon cancer

Dissertation to obtain master degree in Biotechnology

Novel ready-to-eat mango product using gellan gum as gelling agent: physico-chemical, microbial and sensory characteristics

Este trabalho foi efectuado com o apoio da Universidade de Lisboa, Instituto Superior de Agronomia com o Centro de Engenharia dos Biossistemas (CEER

Evaluation of neuroprotective effects of natural extracts obtained from portuguese agro-food residues

Countries are currently faced with problems derived from changes in lifespan and an increase in lifestyle-related diseases. Neurodegenerative disorders such Parkinson’s (PD) and Alzheimer’s (AD) diseases are an increasing problem in aged societies. Data from World Alzheimer Report 2011 indicate that 36 million people worldwide are living with dementia. Oxidative stress has been associated with the development of AD and PD. Therefore there is interest to search for effective compounds or therapies to combat the oxidative damage in these diseases. Current evidence strongly supports a contribution of phenolic compounds present in fruits and vegetables to the prevention of neurodegenerative diseases such AD and PD. The industrial processing of a wide variety of fruits results in the accumulation of by-products without commercial value. Opuntia ficus-indica (cactus pear) is consumed fresh and processed like in juice. Prunnus avium (sweet cherry) is consumed fresh but the organoleptics characteristics of the fruits leads to the smaller and ragged fruits have no commercial value. Fruit extracts of both species has described to be rich in phenolic compounds and to have high antioxidant activities due to its composition. The aim of this work was assessing the efficacy of O. ficus-indica and P. avium by-products extracts obtained with conventional solvent extraction and pressurized liquid extraction in a neurodegeneration cell model. All extracts have protected neuroblastoma cells from H2O2-induced death at low, non-toxic levels, which approach to physiologically-relevant serum concentration. However, cherry extract has a slighter neuroprotective activity. The protective effect of Opuntia extracts are not conducted by a direct antioxidant activity since there are not decreases in intracellular ROS levels in cell treated with extracts and challenged with H2O2, while cherry extract neuroprotection seems to be due to a direct scavenging activity. Extracts from different biological matrixes seems to protect neuronal cells trough different cellular mechanisms.