In this thesis, we developed Bayesian models and machine learning algorithms for protein secondary structure and beta-sheet prediction problems. In protein secondary structure prediction, we developed hidden semi-Markov models, N-best algorithms and training set reduction procedures for proteins in the single-sequence category. We introduced three residue dependency models (both probabilistic and heuristic) incorporating the statistically significant amino acid correlation patterns at structural segment borders. We allowed dependencies to positions outside the segments to relax the condition of segment independence. Another novelty of the models is the dependency to downstream positions, which is important due to asymmetric correlation patterns observed uniformly in structural segments. Among the dataset reduction methods, we showed that the composition based reduction generated the most accurate results. To incorporate non-local interactions characteristic of beta-sheets, we developed two N-best algorithms and a Bayesian beta-sheet model. In beta-sheet prediction, we developed a Bayesian model to characterize the conformational organization of beta-sheets and efficient algorithms to compute the optimum architecture, which includes beta-strand pairings, interaction types (parallel or anti-parallel) and residue-residue interactions (contact maps). We introduced a Bayesian model for proteins with six or less beta-strands, in which we model the conformational features in a probabilistic framework by combining the amino acid pairing potentials with a priori knowledge of beta-strand organizations. To select the optimum beta-sheet architecture, we analyzed the space of possible conformations by efficient heuristics, in which we significantly reduce the search space by enforcing the amino acid pairs that have strong interaction potentials. For proteins with more than six beta-strands, we first computed beta-strand pairings using the BetaPro method. Then, we computed gapped alignments of the paired beta-strands in parallel and anti-parallel directions and chose the interaction types and beta-residue pairings with maximum alignment scores. Accurate prediction of secondary structure, beta-sheets and non-local contacts should improve the accuracy and quality of the three-dimensional structure prediction.

Submitted by Pedro Rafael Costa null (costapr@ibb.unesp.br) on 2016-08-29T12:34:19Z No. of bitstreams: 1 Tese-PedroRafaelCosta.pdf: 3405758 bytes, checksum: db4d115ece49bde1fd0f2a7768c97377 (MD5) / Approved for entry into archive by Ana Paula Grisoto (grisotoana@reitoria.unesp.br) on 2016-08-30T17:27:57Z (GMT) No. of bitstreams: 1 costa_pr_dr_bot.pdf: 3405758 bytes, checksum: db4d115ece49bde1fd0f2a7768c97377 (MD5) / Made available in DSpace on 2016-08-30T17:27:57Z (GMT). No. of bitstreams: 1 costa_pr_dr_bot.pdf: 3405758 bytes, checksum: db4d115ece49bde1fd0f2a7768c97377 (MD5) Previous issue date: 2016-08-23 / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / Uma grande variedade de sequências de RNA presentes nos transcriptomas mas com funções ainda desconhecidas tem estimulado o desenvolvimento de técnicas experimentais e computacionais que colaborem na determinação do papel dessas moléculas. Entretanto, a compreensão dos mecanismos de ação de uma dada molécula de RNA envolve não somente a determinação de sua estrutura de mínima energia livre, mas também o estudo do comportamento de suas conformações metaestáveis. Os algoritmos existentes para predição da estrutura de moléculas de RNA ignoram armadilhas cinéticas que podem levar a formação de estruturas subótimas e utilizam modelos termodinâmicos incompletos, muitas vezes ignorando a formação de estruturas mais complexas, como os pseudonós. Nesse trabalho apresentamos um algoritmo para simulação do dobramento cotranscricional para o estudo dos efeitos da conformação espacial da molécula de RNA na cinética da transcrição. A partir da determinação das conformações permitidas, o algoritmo estabelece uma série de reações de transição entre esses estados, com valores das taxas de ocorrência ponderados através de uma distribuição de probabilidade de Boltzmann baseada na variação de energia livre de Gibbs desses estados. As energias livres das estruturas secundárias são determinadas segundo o modelo dos primeiros vizinhos e dois algoritmos foram implementados para o cálculo da energia livre dos pseudonós. Finalmente, simulações de Monte Carlo baseadas no algoritmo de Gillespie foram realizadas para determinação do caminho de dobramento da molécula. Exemplos de aplicações demonstram o potencial do programa desenvolvido. / The discovery of a wide variety of RNA sequences present in transcriptomes with unknown function has stimulated the development of experimental and computational techniques to determine the function of these molecules. However, understanding the activities of a given RNA molecule involves not only finding its minimum free energy structure, but also studying its metastable structures. The methodologies available for predicting RNA structure ignore kinetic traps that can lead to formation of suboptimal structures and are based in over-simplified thermodynamic models. These methodologies usually can not predict RNA conformations with more complex topologies, such as pseudoknots. In this work we present a computational model for cotranscriptional folding that considers the influence of RNA structures during transcription elongation. After determining the allowed conformations for the sequence of interest, the algorithm established a series of allowed reactions between these structures. The reactions rates are weighted by the Boltzmann factor based on Gibbs free energy variation between the states. The free energies for secondary structures were estimated by the nearest-neighbor model, and two algorithms were implemented to calculate the free energy of pseudoknots. Finally, Monte Carlo simulations based on the Gillespie algorithm were performed to find out the RNA folding path. We show using examples the potential of the software developed. / FAPESP: 2012/19377-4. / CNPq: 152838/2012-0

Orientador: Ney Lemke / Resumo: Uma grande variedade de sequências de RNA presentes nos transcriptomas mas com funções ainda desconhecidas tem estimulado o desenvolvimento de técnicas experimentais e computacionais que colaborem na determinação do papel dessas moléculas. Entretanto, a compreensão dos mecanismos de ação de uma dada molécula de RNA envolve não somente a determinação de sua estrutura de mínima energia livre, mas também o estudo do comportamento de suas conformações metaestáveis. Os algoritmos existentes para predição da estrutura de moléculas de RNA ignoram armadilhas cinéticas que podem levar a formação de estruturas subótimas e utilizam modelos termodinâmicos incompletos, muitas vezes ignorando a formação de estruturas mais complexas, como os pseudonós. Nesse trabalho apresentamos um algoritmo para simulação do dobramento cotranscricional para o estudo dos efeitos da conformação espacial da molécula de RNA na cinética da transcrição. A partir da determinação das conformações permitidas, o algoritmo estabelece uma série de reações de transição entre esses estados, com valores das taxas de ocorrência ponderados através de uma distribuição de probabilidade de Boltzmann baseada na variação de energia livre de Gibbs desses estados. As energias livres das estruturas secundárias são determinadas segundo o modelo dos primeiros vizinhos e dois algoritmos foram implementados para o cálculo da energia livre dos pseudonós. Finalmente, simulações de Monte Carlo baseadas no algoritmo de Gillespie foram ... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: The discovery of a wide variety of RNA sequences present in transcriptomes with unknown function has stimulated the development of experimental and computational techniques to determine the function of these molecules. However, understanding the activities of a given RNA molecule involves not only finding its minimum free energy structure, but also studying its metastable structures. The methodologies available for predicting RNA structure ignore kinetic traps that can lead to formation of suboptimal structures and are based in over-simplified thermodynamic models. These methodologies usually can not predict RNA conformations with more complex topologies, such as pseudoknots. In this work we present a computational model for cotranscriptional folding that considers the influence of RNA structures during transcription elongation. After determining the allowed conformations for the sequence of interest, the algorithm established a series of allowed reactions between these structures. The reactions rates are weighted by the Boltzmann factor based on Gibbs free energy variation between the states. The free energies for secondary structures were estimated by the nearest-neighbor model, and two algorithms were implemented to calculate the free energy of pseudoknots. Finally, Monte Carlo simulations based on the Gillespie algorithm were performed to find out the RNA folding path. We show using examples the potential of the software developed. / Doutor

Orientadores: Ana Maria Lima de Azeredo-Espin, Nilson Ivo Tonin Zanchin / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Biologia / Made available in DSpace on 2018-08-19T22:51:23Z (GMT). No. of bitstreams: 1 Marinho_MarcoAntonioTonus_D.pdf: 17816653 bytes, checksum: e915a871c22b7741c1be765f87c95ff5 (MD5) Previous issue date: 2011 / Resumo: A superfamília Oestroidea (Diptera:Brachycera:Calyptratae), com +13.000 espécies descritas, compreende um dos grupos mais numerosos e ecologicamente diversos da ordem Diptera. O grupo possui grande interesse para atividades humanas por englobar espécies de importância médica, veterinária e forense, muitas das quais compõem a família Calliphoridae. Apesar do grande número de estudos disponíveis, as relações evolutivas no grupo, o qual é composto predominantemente por linhagens de rápida diversificação e radiação, ainda são controversas e pouco compreendidas, encorajando a caracterização de novos marcadores moleculares para análises de filogenia molecular. Neste contexto, esta tese foi desenvolvida e organizada em três capítulos descrevendo estudos genéticoevolutivos em espécies da superfamília Oestroidea, com ênfase em Calliphoridae. O primeiro capítulo trata da caracterização e avaliação do segundo espaçador transcrito interno (ITS2) do DNA ribossomal como um marcador molecular para análises filogenéticas em Calliphoridae, incorporando informações tanto da sequência primária quanto da estrutura secundária adquirida pela região. A análise do ITS2 revelou um padrão hierarquicamente organizado das distâncias genéticas nos níveis de espécies, gêneros e subfamílias, enquanto pouca variação intra-específica foi encontrada. As árvores inferidas recuperaram muitas das relações comumente aceitas entre os táxons amostrados, sendo que a inclusão da informação estrutural nas análises resultou na recuperação de topologias mais confiáveis. Sendo assim, o potencial da região ITS2 como um marcador molecular para análises evolutivas na família Calliphoridae foi confirmado e seu uso em análises de maior escala, incluindo marcadores de diferentes naturezas de evolução, encorajado. O segundo capítulo da tese descreve a caracterização in vitro da estrutura secundária adquirida pelo ITS2, através de padrões de digestão enzimática e análise dos fragmentos gerados, em espécies representantes das três superfamílias de Calyptratae: Glossina morsitans, Musca domestica e Cochliomyia hominivorax. A análise do padrão de fragmentos gerados pelas enzimas RNAse I, A, T1 e V1, quando mapeados na estrutura secundária predita in silico, corroborou muitos dos domínios inicialmente preditos pelo método computacional, ressaltando a importância e confiabilidade desses métodos na predição de estruturas secundárias. O terceiro capítulo da tese descreve análises de filogenia molecular na superfamília Oestroidea, com ênfase na amostragem de espécies de Calliphoridae, utilizando quatro marcadores moleculares, dois nucleares (ITS2 e 28S) e dois mitocondriais (COI e 16). As análises, que incluíram uma extensa avaliação dos efeitos de diferentes estratégias de particionamento dos dados em análises de inferência Bayesiana (por conformação estrutural e posição no códon), revelaram a existência de dois clados principais em Oestroidea: Tachinidae + Mesembrinellinae e Oestridae + Rhiniinae + Sarcophagidae + Calliphoridae (definida em senso estrito). O status de família recentemente atribuído à Rhiniinae foi encontrado, enquanto há também evidências para sugerir o mesmo para a subfamília Mesembrinellinae, como proposto anteriormente por outros autores. As diferentes estratégias de particionamento do conjunto de dados amostrados resultaram em diferenças discretas em termos de topologia, comprimentos de ramo e suporte geral das filogenias inferidas. Embora o resultado geral indique uma melhor resolução das análises quando do uso de combinações de partições e modelos mais complexas, as mesmas podem ocasionar também um aumento considerável na incerteza associada às análises / Abstract: The Oestroidea superfamily (Diptera: Brachycera: Calyptratae), with +13,000 described species, comprises one of the most numerous and ecologically diverse groups in the Diptera order. The group is actually of great interest for human activities since it includes species of medical, veterinary and forensic importance, most of them included in the Calliphoridae family. Despite the existence of several studies addressing the issue, evolutionary relationships in Oestroidea, a group mainly composed of rapidly diverged lineages, remains contentious and poorly understood, encouraging the characterization of new molecular markers for phylogenetic inference analyses. In this context, this thesis was developed and organized in three chapters describing genetic and evolutionary studies in species of the Oestroidea superfamily, with emphasis in Calliphoridae. The first chapter deals with the characterization and evaluation of the second internal transcribed spacer region (ITS2) of the ribosomal DNA cluster as a molecular marker for phylogenetic inference in Calliphoridae, including information of both primary sequence and secondary structure. The analyses revealed an hierarchically organized pattern of genetic distances in the specific, generic and subfamilial level, while little intraspecific variation was detected. Inferred trees were able to recover most of the commonly accepted relationships among the sampled taxa, with the consideration of structural information resulting in better supported topologies. Thereby, the potential of the ITS2 region as a molecular marker for phylogenetic inference in the Calliphoridae family was corroborated and its use in larger scale analyses, including other markers with different evolutionary patterns, encouraged. Chapter II describes the in vitro characterization of the secondary structure of the ITS2 region, through patterns of enzymatic digestion and analysis of the generated fragments, in representative species of the three superfamilies of the Calyptratae clade: Glossina morsitans, Musca domestica and Cochliomyia hominivorax. Analyses of the patterns of the fragments generated by enzymatic digestions with the RNAses I, A, T1 and V1, when mapped in the in silico predicted secondary structure, corroborated the folding of most of the domains predicted by computational methods, highlighting the importance and reliability of these methods in secondary structure prediction. Chapter III describes molecular phylogenetic analyses in the Oestroidea superfamily, with emphasis on the Calliphoridae family, using four different molecular markers, two nuclear (ITS2 and 28S) and two mitochondrial (COI and 16S) regions. The analyses, which included a comprehensive evaluation of the effects of different data partitioning strategies in a Bayesian framework (by structural conformation and codon position), revealed the existence of two main clades in Oestroidea: Tachinidae+Mesembrinellinae and Oestridae+Rhiniinae+Sarcophagidae+Calliphoridae (defined in a strict sense). The recently attributed family status to Rhiniinae was confirmed, and there are evidence to also suggest the same for Mesembrinellinae, as previously pointed out by other studies. The different data partitioning strategies used in the sampled dataset resulted in small differences in terms of inferred topologies, estimated branch lengths and average support. Although the overall results indicate a significant increase in phylogeny resolution when more complex and parameter-rich models / partitions combinations are used, they can also lead to an increased uncertainty in the phylogenetic estimation process / Doutorado / Genetica Animal e Evolução / Soutor em Genética e Biologia Molecular

Since the time RNA has been discovered by the nature scientist Miescher the structure and function of it has been forgotten for a long time. The prime role in science had always DNA. An increase of interest in RNA came with the discovery of the tRNA structure and its catalytic and enzymatic properties. These discoveries led to a great development wave of bioinformatics and structure and function analysis of RNA.

Magister Scientiae - MSc / In summary there currently exist techniques to discover miRNA however both require many calculations to be performed during the identification limiting their use at a genomic level. Machine learning techniques are currently providing the best results by combining a number of calculated and statistically derived features to identify miRNA candidates, however almost all of these still include computationally intensive secondary-structure calculations. It is the aim of this project to produce a miRNA identification process that minimises and simplifies the number of computational elements required during the identification process. / South Africa

One of the challenges in the structural biological community is processing the wealth of protein data being produced today; therefore, the use of computational tools has been incorporated to speed up and help understand the structures of proteins, hence the functions of proteins. In this thesis, protein structure investigations were made through the use of Multivariate Analysis (MVA), and Fourier Transformed Infrared (FTIR), a form of vibrational spectroscopy. FTIR has been shown to identify the chemical bonds in a protein in solution and it is rapid and easy to use; the spectra produced from FTIR are then analysed qualitatively and quantitatively by using MVA methods, and this produces non-redundant but important information from the FTIR spectra. High resolution techniques such as X-ray crystallography and NMR are not always applicable and Fourier Transform Infrared (FTIR) spectroscopy, a widely applicable analytical technique, has great potential to assist structure analysis for a wide range of proteins. FTIR spectral shape and band positions in the Amide I (which contains the most intense absorption region), Amide II, and Amide III regions, can be analysed computationally, using multivariate regression, to extract structural information. In this thesis Partial least squares (PLS), a form of MVA, was used to correlate a matrix of FTIR spectra and their known secondary structure motifs, in order to determine their structures (in terms of "helix", "sheet", “310-helix”, “turns” and "other" contents) for a selection of 84 non-redundant proteins. Analysis of the spectral wavelength range between 1480 and 1900 cm-1 (Amide I and Amide II regions) results in high accuracies of prediction, as high as R2 = 0.96 for α-helix, 0.95 for β-sheet, 0.92 for 310-helix, 0.94 for turns and 0.90 for other; their Root Mean Square Error for Calibration (RMSEC) values are between 0.01 to 0.05, and their Root Mean Square Error for Prediction (RMSEP) values are between 0.02 to 0.12. The Amide II region also gave results comparable to that of Amide I, especially for predictions of helix content. We also used Principal Component Analysis (PCA) to classify FTIR protein spectra into their natural groupings as proteins of mainly α-helical structure, or protein of mainly β-sheet structure or proteins of some mixed variations of α-helix and β-sheet. We have also been able to differentiate between parallel and anti-parallel β-sheet. The developed methods were applied to characterize the secondary structure conformational changes of an unfolding protein as a function of pH and also to determine the limit of Quantitation (LoQ).Our structural analyses compare highly favourably to those in the literature using machine learning techniques. Our work proves that FTIR spectra in combination with multivariate regression analysis like PCA and PLS, can accurately identify and quantify protein secondary structure. The developed models in this research are especially important in the pharmaceutical industry where the therapeutic effect of drugs strongly depends on the stability of the physical or chemical structure of their proteins targets; therefore, understanding the structure of proteins is very important in the biopharmaceutical world for drugs production and formulation. There is a new class of drugs that are proteins themselves used to treat infectious and autoimmune diseases. The use of spectroscopy and multivariate regression analysis in the medical industry to identify biomarkers in diseases has also brought new challenges to the bioinformatics field. These methods may be applicable in food science and academia in general, for the investigation and elucidation of protein structure.

This master's thesis focuses on search and structure prediction of quadruplexes in DNA sequences. Thesis also explains related terms that are important for understanding the function, properties and geometry of quadruplexes. Thesis describe physico-chemical and computational current methods, which possible to discover and structure prediction. This paper also explain the principle of molecular modelling, which was used in the final application. Design and implementation of the final algorithm are also part of this thesis.

This master’s thesis deals with protein secondary structure prediction. There is a theoretical introduction followed by study of available tools, proposal and implementation of web application, which combines functionality of several web tools used to predict secondary structure. User is asked to choose prediction methods and insert input sequence as plain text or upload a file. Results collected from selected tools serve to convert data into common format, show the result and create new type of prediction. Finally, the testing is applied and influences of tools are adjusted in order to increase percentage of prediction. The output of application is a result of prediction also available as plain text or as a file.

<p>The hydrogen bond is one of nature’s ubiquitous molecular interactions. Its role ranges from that of a static provider of structural integrity in proteins to that of a dynamic coordinate, along which excited state deactivation in sunscreen molecules is achieved. The work in this dissertation employs a supersonic expansion to collisionally cool peptide oligomers and a sunscreen chromophore to the zero-point vibrational level of their low lying conformational minima. These species are interrogated using high-resolution, conformer-specific ultraviolet and infrared laser spectroscopic techniques with the aim of elucidating their intrinsic conformational preferences, hydrogen bonding networks, and excited state deactivation mechanisms.</p><p>Synthetic foldamers are oligomers composed of non-natural building blocks, such as b- and g-amino acids. Incorporation of such residues into a peptide backbone results in secondary and tertiary structures that are distinct from those found in nature. Herein, the folding propensity of a series of mixed a/b and pure b-peptides is presented. In each case, both the left- and right-handed emergence of mixed-helical secondary structures, the 11/9- and the 12/10-helix, are observed. Next, the intrinsic conformational preferences of a series of increasingly complex asparagine-containing peptides are characterized. Asparagine, with its flexible carboxamide sidechain, is omnipresent within the prion forming domain of the misfolded proteins associated with several neurodegenerative diseases. Asparagine’s propensity for b-turn structures is discussed and compared with that of analogous peptide sequences found in nature.</p><p>Methyl anthranilate is a natural product that contains an identical electronic chromophore to the sunscreen agent, meradimate. The excited state deactivation mechanism of methyl anthranilate and its water complex is determined with extensive ultraviolet spectroscopic characterization, and is discussed within the broader context of its role as a sunscreen agent. Vibronic analysis coupled with computational results indicate extensive heavy-atom rearrangement leading to hydrogen atom dislocation, rather than full transfer, on the S₁ surface. This phenomenon is further characterized with infrared spectroscopy and theoretical modeling, in which the NH stretch is adiabatically separated from other internal coordinates. Extensive dilution of the dislocated NH stretch oscillator strength over many transitions and ~1,300 cm^-1 is predicted. These results may have implications for similar molecules, such as salicylic acid and its derivatives.</p>