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.

Reactive electron scattering from biomolecules and technologically relevant molecules

The role of a set of gases relevant within the context of biomolecules and technologically relevant molecules under the interaction of low-energy electrons was studied in an effort to contribute to the understanding of the underlying processes yielding negative ion formation. The results are relevant within the context of damage to living material exposed to energetic radiation, to the role of dopants in the ion-molecule chemistry processes, to Electron Beam Induced Deposition (EBID) and Ion Beam Induced Deposition (IBID) techniques. The research described in this thesis addresses dissociative electron attachment (DEA) and electron transfer studies involving experimental setups from the University of Innsbruck, Austria and Universidade Nova de Lisboa, Portugal, respectively. This thesis presents DEA studies, obtained by a double focusing mass spectrometer, of dimethyl disulphide (C2H6S2), two isomers, enflurane and isoflurane (C3F5Cl5) and two chlorinated ethanes, pentachloroethane (C2HCl5) and hexachloroethane (C2Cl6), along with quantum chemical calculations providing information on the molecular orbitals as well as thermochemical thresholds of anion formation for enflurane, isoflurane, pentachloroethane and hexachloroethane. The experiments represent the most accurate DEA studies to these molecules, with significant differences from previous work reported in the literature. As far as electron transfer studies are concerned, negative ion formation in collisions of neutral potassium atoms with N1 and N3 methylated pyrimidine molecules were obtained by time-of-flight mass spectrometry (TOF). The results obtained allowed to propose concerted mechanisms for site and bond selective excision of bonds.

Biotoxicity assays of quantum dots in in vitro cultures of Medicago sativa and Medicago truncatula

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

Effect of fungal colonization on mechanical performance of cork

The industrialization of traditional processes relies on the scientific ability to understand the empirical evidence associated with traditional knowledge. Cork manufacturing includes one operation known as stabilization, where humid cork slabs are extensively colonized by fungi. The implications of fungal growth on the chemical quality of cork through the analysis of putative fungal metabolites have already been investigated. However, the effect of fungal growth on the mechanical properties of cork remains unexplored. This study investigated the effect of cork colonization on the integrity of the cork cell walls and their mechanical performance. Fungal colonization of cork by Chrysonilia sitophila, Mucor plumbeus Penicillium glabrum, P. olsonii, and Trichoderma longibrachiatum was investigated by microscopy. Growth occurred primarily on the surface of the cork pieces, but mycelium extended deeper into the cork layers, mostly via lenticular channels and by hyphal penetration of the cork cell wall. In this first report on cork decay in which specific correlation between fungal colonization and mechanical proprieties of the cork has been investigated, all colonizing fungi except C. sitophila, reduced cork strength, markedly altering its viscoelastic behaviour and reducing its Young’s modulus.

The impact of CdSe/ZnS Quantum Dots in cells of Medicago sativa in suspension culture

Abstract Background: Nanotechnology has the potential to provide agriculture with new tools that may be used in the rapid detection and molecular treatment of diseases and enhancement of plant ability to absorb nutrients, among others. Data on nanoparticle toxicity in plants is largely heterogeneous with a diversity of physicochemical parameters reported, which difficult generalizations. Here a cell biology approach was used to evaluate the impact of Quantum Dots (QDs) nanocrystals on plant cells, including their effect on cell growth, cell viability, oxidative stress and ROS accumulation, besides their cytomobility. Results: A plant cell suspension culture of Medicago sativa was settled for the assessment of the impact of the addition of mercaptopropanoic acid coated CdSe/ZnS QDs. Cell growth was significantly reduced when 100 mM of mercaptopropanoic acid -QDs was added during the exponential growth phase, with less than 50% of the cells viable 72 hours after mercaptopropanoic acid -QDs addition. They were up taken by Medicago sativa cells and accumulated in the cytoplasm and nucleus as revealed by optical thin confocal imaging. As part of the cellular response to internalization, Medicago sativa cells were found to increase the production of Reactive Oxygen Species (ROS) in a dose and time dependent manner. Using the fluorescent dye H2DCFDA it was observable that mercaptopropanoic acid-QDs concentrations between 5-180 nM led to a progressive and linear increase of ROS accumulation. Conclusions: Our results showed that the extent of mercaptopropanoic acid coated CdSe/ZnS QDs cytotoxicity in plant cells is dependent upon a number of factors including QDs properties, dose and the environmental conditions of administration and that, for Medicago sativa cells, a safe range of 1-5 nM should not be exceeded for biological applications.

Modelling of the kinetics of municipal solid waste composting in full-scale mechanical-biological treatment plants

A thesis submitted in fulfilment of the requirements for the Degree of Doctor of Philosophy in Sanitary Engineering in the Faculty of Sciences and Technology of the New University of Lisbon

Study of forbidden transitions in atomic systems

A Thesis submitted for the co-tutelle degree of Doctor in Physics at Universidade Nova de Lisboa and Université Pierre et Marie Curie

Probabilistic models for mechanical properties of prestressing strands

This study focus on the probabilistic modelling of mechanical properties of prestressing strands based on data collected from tensile tests carried out in Laboratório Nacional de Engenharia Civil (LNEC), Portugal, for certification purposes, and covers a period of about 9 years of production. The strands studied were produced by six manufacturers from four countries, namely Portugal, Spain, Italy and Thailand. Variability of the most important mechanicalproperties is examined and the results are compared with the recommendations of the ProbabilisticModel Code, as well as the Eurocodes and earlier studies. The obtained results show a very low variability which, of course, benefits structural safety. Based on those results, probabilistic modelsfor the most important mechanical properties of prestressing strands are proposed.

Correlation between the mechanical cycling behavior and microstructure in laser welded joints using NiTi memory shape alloys

Dissertação para obtenção do Grau de Mestre em Engenharia de Materiais

Ligand K-edge X-ray absorption spectroscopy and DFT calculations on [Fe3S4]0,+ clusters: delocalization, redox, and effect of the protein environment

Ligand K-edge XAS of an [Fe3S4]0 model complex is reported. The pre-edge can be resolved into contributions from the í2Ssulfide, í3Ssulfide, and Sthiolate ligands. The average ligand-metal bond covalencies obtained from these pre-edges are further distributed between Fe3+ and Fe2.5+ components using DFT calculations. The bridging ligand covalency in the [Fe2S2]+ subsite of the [Fe3S4]0 cluster is found to be significantly lower than its value in a reduced [Fe2S2] cluster (38% vs 61%, respectively). This lowered bridging ligand covalency reduces the superexchange coupling parameter J relative to its value in a reduced [Fe2S2]+ site (-146 cm-1 vs -360 cm-1, respectively). This decrease in J, along with estimates of the double exchange parameter B and vibronic coupling parameter ì2/k-, leads to an S ) 2 delocalized ground state in the [Fe3S4]0 cluster. The S K-edge XAS of the protein ferredoxin II (Fd II) from the D. gigas active site shows a decrease in covalency compared to the model complex, in the same oxidation state, which correlates with the number of H-bonding interactions to specific sulfur ligands present in the active site. The changes in ligand-metal bond covalencies upon redox compared with DFT calculations indicate that the redox reaction involves a two-electron change (one-electron ionization plus a spin change of a second electron) with significant electronic relaxation. The presence of the redox inactive Fe3+ center is found to decrease the barrier of the redox process in the [Fe3S4] cluster due to its strong antiferromagnetic coupling with the redox active Fe2S2 subsite.

Quantum dots: synthesis, functionalization and bioconjugation for biological applications

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