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.

Wetland Habitat Studies using various Classification Techniques on Multi-Spectral Landsat Imagery: Case study: Tram chim National Park, Dong Thap Vietnam

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

Automatic sleep apnea detection and sleep classification using the ECG and the SpO2 signals

Dissertation for a Masters Degree in Computer and Electronic Engineering

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

The contribution of multitemporal information from multispectral satellite images for automatic land cover classification at the national scale

Thesis submitted to the Instituto Superior de Estatística e Gestão de Informação da Universidade Nova de Lisboa in partial fulfillment of the requirements for the Degree of Doctor of Philosophy in Information Management – Geographic Information Systems

Unsupervised automatic music genre classification

Trabalho apresentado no âmbito do Mestrado em Engenharia Informática, como requisito parcial para obtenção do grau de Mestre em Engenharia Informática

Automatic cymbal classification

Dissertação apresentada na Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa para a obtenção do grau de Mestre em Engenharia Informática

Multi-scale vegetation classification using earth observation data of the Sundarban mangrove forest, Bangladesh.

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

Dynamic and interaction of cytochrome c with Pf1 virus

Dissertação apresentada na Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa para obtenção do grau de Mestre em BioOrgânica

Strategies of the African swine fever virus to manipulate innate immunity

Dissertation presented to obtain the Ph.D degree in Biology

Expressão em Escherichia coli de antigénios do Cell fusing agent virus (Flaviviridae: Flavivirus) como proteína de fusão

RESUMO: O Cell Fusing Agent Vírus (CFAV), considerado como o primeiro “flavivírus específicos de insectos” (ISF), parece estar exclusivamente adaptado aos seus hospedeiros, não replicando em células de vertebrados. Apesar de ter sido identificado há mais de três décadas (1975), a verdade é que muito pouco se conhece sobre a sua biologia. Dado o seu parentesco filogenético com alguns outros flavivírus encontrados naturalmente em mosquitos de diferentes géneros colhidos em diferentes regiões do globo, este vírus poderá ser usado como modelo para o estudo de ISF. No entanto, necessitam do desenvolvimento de ferramentas básicas, tais como clones moleculares ou baterias de soros contendo anticorpos que reconheçam uma ou mais proteínas codificadas pelo genoma viral, produzidas, por exemplo, a partir de antigénios virais produzidos de forma recombinante. Com este trabalho pretendeu-se a optimização de protocolos que permitiram a expressão e purificação parcial de quatro proteínas [duas proteínas estruturais (C e E) e duas não estruturais (NS3hel e NS5B)] do CFAV em E. coli, todas elas produzidas como proteínas de fusão com “caudas” (tags) de hexahistidina nos seus extremos carboxilo. Para a expansão do CFAV foram utilizadas células Aedes albopictus (C6/36). Após a realização da extracção do RNA viral e a obtenção de cDNA, procedeu-se amplificação, por RT-PCR, das regiões codificantes das proteínas C, E, NS3hel e NS5B, utilizando primers específicos. Os quatro fragmentos de DNA foram independentemente inseridos no vector pJTE1.2/blunt usando E. coli NovaBlue como hospedeira de clonagem e, posteriormente, inseridos em vectores de expressão pET-28b e pET-29b usando E. coli BL21(DE3)pLysS e Rosetta(DE3)pLysS como hospedeiras de expressão. Após da indução, expressão e purificação das proteínas recombinantes C, E, NS3hel e NS5B, foi confirmada a autenticidade destas proteínas produzidas através do método Western Blot com um anticorpo anti-histidina. --------- ABSTRACT: The Cell Fusing Agent virus (CFAV) considered as the first "insect- specific flavivirus" (ISF) and seems to be uniquely adapted to their hosts, not replicating in vertebrate cells. Although it has been known for more than three decades (1975), the truth is very little is known about its biology. Given its close phylogenetic relationship with other flavivirus naturally circulating in various genera of mosquitoes collected from different regions of the globe, this virus could be used as a model for the study of ISF. However, such studies require the development of experimental basic tools, such as molecular clones or serum batteries containing antibodies that recognize one or more proteins encoded by the viral genome, produced, for example, from viral antigens recombinant produced. In this work, we carried out the optimization of protocols that allowed the expression and partial purification of four proteins [two structural proteins (C and E) and two nonstructural proteins (NS3hel and NS5B)] CFAV in E. coli as fusion protein for c-terminal hexahistidine tags. For the expansion of the CFAV we used Aedes albopictus (C6/36) cells. After completion of the viral RNA extraction and cDNA obtained, amplification of the coding regions of the C, E, NS5B and NS3hel proteins was carried out by RT-PCR using specific primers. The four DNA fragments were independently inserted into the vector pJTE1.2/blunt using E. coli NovaBlue as cloning host and then inserted into expression vectors pET-28b and pET-29b using E. coli BL21(DE3)pLysS and Rosetta(DE3)pLysS as expression host. After induction, expression and purification of recombinant C, E, NS3hel and NS5B proteins Western Blot analyses with an anti-histidine antibody confirmed the authenticity of these proteins produced.

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.

Contract-program evaluation: Impact of the DRG classification system on hospitals' outputs

A Work Project, presented as part of the requirements for the Award of a Masters Degree in Management from the NOVA – School of Business and Economics

Integrated functional genomics for improved manufacture of recombinant enveloped virus

Dissertation presented to obtain the Ph.D degree in Engineering and Technology Sciences-Biotechnology

Exploratory analysis of OpenStreetMap for land use classification

In the last years, volunteers have been contributing massively to what we know nowadays as Volunteered Geographic Information. This huge amount of data might be hiding a vast geographical richness and therefore research needs to be conducted to explore their potential and use it in the solution of real world problems. In this study we conduct an exploratory analysis of data from the OpenStreetMap initiative. Using the Corine Land Cover database as reference and continental Portugal as the study area, we establish a possible correspondence between both classification nomenclatures, evaluate the quality of OpenStreetMap polygon features classification against Corine Land Cover classes from level 1 nomenclature, and analyze the spatial distribution of OpenStreetMap classes over continental Portugal. A global classification accuracy around 76% and interesting coverage areas’ values are remarkable and promising results that encourages us for future research on this topic.