Dissertations / Theses on the topic 'Bioinformatique structurale'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Bioinformatique structurale.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
Madaoui, Hocine. "Prédiction structurale et ingénierie des assemblages macromoléculaires par bioinformatique." Phd thesis, Université Paris-Diderot - Paris VII, 2007. http://tel.archives-ouvertes.fr/tel-00553875.
Full textMadaoui, Hocine. "Prédiction structurale et ingenierie des assemblages macromoléculaires par bioinformatique." Paris 7, 2007. https://tel.archives-ouvertes.fr/tel-00553875.
Full textThe high-throughput characterization of the protein-protein interactions networks laid the bases for the first interaction maps in all model organisms, including human. In contrast, the structures of the protein assembles are still restricted to a very limited set of interactions. In this work, a specific evolutionary pressure that is exerted at protein interfaces has been revealed. To our knowledge, no such effect had been previously described. Based on this finding, a novel bioinformatic approach, called scotch (surface complementarity trace in complex history) has been developed to predict the structures of protein assembles. Coupled to a docking program, such as scotcher also developed in this work, this approach was shown to predict efficiently the structures of many complexes. This work also focuses on the inhibition of protein interactions by synthetic peptides, rationally designed on the basis of the complex structure. The results obtained for two examples, the asf1 - histone h3/h4 and the gp120 - cd4 complexes emphasize the high interest of rational design of complex interface for the development of novel therapeutic strategies
Sagot, Marie-France. "Ressemblance lexicale et structurale entre macromolécules : formalisation et approches combinatoires." Marne-la-Vallée, 1996. http://www.theses.fr/1996MARN0049.
Full textPihan, Émilie. "Recherche de nouveaux antipaludiques par bioinformatique structurale et chémoinformatique : application à deux cibles : PfAMA1 et PfCCT." Thesis, Nice, 2013. http://www.theses.fr/2013NICE4039/document.
Full textHuman malaria is caused by five parasitic species of the genus Plasmodium, P. falciparum being the most deadly. Drug resistance of some parasite strains has been reported for commercial drugs. Vector mosquitoes are resistant to perythroid insecticides and no successful vaccine is available. This disease is a public and economic health issue for developing countries. My PhD projects investigate new treatments for malaria, by targeting two new proteins. Apicomplexa parasites have developed a unique invasion mechanism involving a tight interaction formed between the host cell and the parasite surfaces called Moving Junction. The structural and functional characterization of the AMA1-RON2 complex pave the way for the design of low molecular weight compounds capable of disrupting the AMA1-RON2 assembly and thereby invasion. The parasite also needs phospholipids to build its membrane during the erythrocytic cycle. There are six times more phospholipids in infected erythrocytes compared to healthy ones. Our strategy is to inhibit the de novo Kennedy pathway and more precisely its rate-limiting step catalysed by the enzyme PfCCT. Filters were used for ligand-based (LBVS) and structure-based virtual screening (SBVS) of commercial chemical databases that I have prepared. For each project, molecules were selected in terms of their docking scores and their interactions with key active site residues. By combining structural bioinformatics and cheminformatics, we identified potential inhibitors of the two protein targets
Pihan, Émilie. "Recherche de nouveaux antipaludiques par bioinformatique structurale et chémoinformatique : application à deux cibles : PfAMA1 et PfCCT." Electronic Thesis or Diss., Nice, 2013. http://www.theses.fr/2013NICE4039.
Full textHuman malaria is caused by five parasitic species of the genus Plasmodium, P. falciparum being the most deadly. Drug resistance of some parasite strains has been reported for commercial drugs. Vector mosquitoes are resistant to perythroid insecticides and no successful vaccine is available. This disease is a public and economic health issue for developing countries. My PhD projects investigate new treatments for malaria, by targeting two new proteins. Apicomplexa parasites have developed a unique invasion mechanism involving a tight interaction formed between the host cell and the parasite surfaces called Moving Junction. The structural and functional characterization of the AMA1-RON2 complex pave the way for the design of low molecular weight compounds capable of disrupting the AMA1-RON2 assembly and thereby invasion. The parasite also needs phospholipids to build its membrane during the erythrocytic cycle. There are six times more phospholipids in infected erythrocytes compared to healthy ones. Our strategy is to inhibit the de novo Kennedy pathway and more precisely its rate-limiting step catalysed by the enzyme PfCCT. Filters were used for ligand-based (LBVS) and structure-based virtual screening (SBVS) of commercial chemical databases that I have prepared. For each project, molecules were selected in terms of their docking scores and their interactions with key active site residues. By combining structural bioinformatics and cheminformatics, we identified potential inhibitors of the two protein targets
Magis, Cedrik. "Conception de Ligands Protéiques par Bioinformatique et Modélisation Moléculaire." Phd thesis, Museum national d'histoire naturelle - MNHN PARIS, 2007. http://tel.archives-ouvertes.fr/tel-00553476.
Full textNehdi, M. Atef. "Étude structurale du ribozyme VHD antigénomique par évolution in vitro couplée à une analyse bioinformatique." Thèse, Université de Sherbrooke, 2007. http://savoirs.usherbrooke.ca/handle/11143/4241.
Full textNehdi, M. Atef. "Étude structurale du ribozyme VHD antigénomique par évolution in vitro couplée à une analyse bioinformatique." [S.l. : s.n.], 2007.
Find full textAndreani, Jessica. "Analyse évolutive, prédiction structurale et inhibition des interactions protéine-protéine." Paris 6, 2013. http://www.theses.fr/2013PA066291.
Full textLes interactions protéine-protéine sont fondamentales dans la plupart des processus cellulaires. Cette thèse est centrée sur l’analyse et la prédiction de ces interactions en utilisant à la fois les données structurales et l’information issue de l’évolution. A travers l’étude de plus de 1000 couples d’interfaces homologues, extraits d’une base de données développée dans notre équipe, nous avons mis en évidence une plasticité étonnante dans l’évolution de la structure des interfaces. Nous avons cependant identifié des propriétés assez conservées qui fournissent des pistes pour l’extraction d’information à partir des alignements de séquences multiples de deux partenaires en interaction. Nous avons ensuite développé une fonction de score « gros grain » utilisant un potentiel statistique multi-corps couplé à l’information évolutive. Cette fonction améliore les prédictions d’interfaces protéiques et a été utilisée dans deux cas concrets d’amarrage moléculaire. Enfin, nous avons développé un protocole bio-informatique robuste pour le design d’inhibiteurs peptidiques d’une interaction protéine-protéine
Desmet, François-Olivier. "Bioinformatique et épissage dans les pathologies humaines." Thesis, Montpellier 1, 2010. http://www.theses.fr/2010MON1T017.
Full textDiscovered in 1977, splicing is a post-transcriptional maturation process that consists in link-ing exons together and removing introns from a pre-messanger RNA. For splicing to be cor-rectly undertaken by the spliceosome and its auxiliary proteins, several signals are located along the pre-messanger RNA sequence. Nearly half of pathogenous mutations in humans are now recognized to impact splicing and leading to a gene dysfunction. Therefore it is es-sential for biologists to detect those signals in any genomic sequence.Thus, the goals of this thesis were to conceive new algorithms: i) to identify splicing signals; ii) to predict the impact of mutations on these signals and iii) to give access to this information to researchers thanks to the power of bioinformatics. The proposed solution, Human Splicing Finder (HSF), is a web application able to predict all types of splicing signals hidden in any sequence extracted from the human genome. We demonstrated the prediction's efficiency of HSF for all situations associated with pathogenous mutations for which an impact on splicing has been experimentally demonstrated. Along with these direct benefits for the knowledge of biological processes for splicing and diagnosis, new genotype-specific therapeutic approaches can also benefit from these new algorithms. Thus, HSF allows to better target antisense olignucleotides used to induce exon skipping in Duchenne myopathy and dysferlinopathies.The recent recognition of the major interest of splicing in various domains such as fundamen-tal research, therapeutics and diagnosis needed a one stop shop for splicing signals. HSF has for object to fulfill this need, being regularly updated to integrate new knowledge and is already recognized as an international reference tool
Lallous, Nada. "Étude structurale et fonctionnelle des modules de reconnaissance des marques épigénétiques dans la protéine humaine UHRF1." Strasbourg, 2010. http://www.theses.fr/2010STRA6017.
Full textThe human UHRF1 protein, also called ICBP90 (Inverted CCAAT box Binding protein of 90 kDa), is a multi-domain nuclear protein able to recognize different epigenetic marks and to interact with different actors of the epigenetic regulation. During this work, we characterized structurally and functionally the SRA (Set and RING associated domain) and the PHD (Plant Homeodomain) domains of hUHRF1, known to interact with hemimethylated DNA and histone H3 respectively. Different biophysical methods were used to characterize the associated domains PHD and SRA, to help in understanding the biological function of these two chromatin recognition modules in the UHRF1 protein. The three-dimensional structures of the hUHRF1 PHD domain, alone and in complex with a histone H3 peptide, were determined by X-ray crystallography and the specificity of histone H3 recognition by the PHD domain was characterized in solution
Piuzzi, Marc. "Détermination de la structure de protéines à l’aide de données faiblement résolues." Paris 6, 2010. http://www.theses.fr/2010PA066510.
Full textGaschignard, Geoffroy. "Étude structurale de la calcyanine, nouvelle protéine impliquée dans la biominéralisation intracellulaire chez les cyanobactéries." Electronic Thesis or Diss., Sorbonne université, 2023. http://www.theses.fr/2023SORUS272.
Full textBiomineralisation is all the processes that lead to the formation of minerals by living beings. In 2012, a new biomineralization phenotype has been described in cyanobacteria, characterized by the presence of amorphous alkaline-earth carbonate inclusions inside the cells. A comparative genomic analysis revealed that this intracellular biomineralisation phenotype is linked to the presence of one gene, unknown at the time, which has been called ccyA. It codes for one protein called calcyanin. Calcyanin has 4 variants, that share the same C-terminus domain ((GlyZip)3), but which differ in their N-terminus domain (CoBaHMA, X, Y or Z). None of these 5 domains has already been described in the literature. The goal of this PhD was to characterize the 3D structure of Synechococcus calcipolaris’s calcyanin, which has a CoBaHMA domain, by combining bioinformatics and experimental approaches, in order to make hypothesis regarding its role. Through sequence analysis and 3D structure modeling, we showed that the CoBaHMA domain belongs to the “ferredoxin-like” fold, typical of the superfamily HMA (“Heavy Metal Associated”), and sets itself as a new family in it, characterized by conserved basic amino acids and an additional β strand (1). We have performed sequence similarity searches, refined with the structural information of the 3D structure models. This way, we showed that the CoBaHMA domain can be found on several different protein architectures, in various taxa. It exists has an independent domain, or in conjunction with other domains, especially membrane systems which, among others, allow transports of substrates through the membrane (PIB-type ATPases, ABC exporters) or new families with unknown functions. These results lead us to formulate hypotheses regarding the CoBaHMA domain function (2). We also proposed a robust model for the individual glycine zipper from which the name of the C-terminus domain (GlyZip)3 of calcyanins comes from. These glycine zippers have a structure of a compact hairpin made of two helices, which is akin to the ones of transmembrane proteins that form pore. However, we were not able to model satisfyingly their assembly nor their possible interactions with the CoBaHMA domain, emphasizing the importance of studying the protein experimentally. We successfully expressed calcyanin in Eschericha coli, and purified it. However the protein proved to be quite unstable, with a propensity to form a great diversity of objects with different sizes. A limited proteolysis experiment revealed the existence of a protease-resistant fragment of calcyanin, which encompasses the CoBaHMA domain and the first glycine zipper of the (GlyZip)3 domain. We expressed and purified this fragment, fused to MBP (« Maltose Binding Protein »). The fragment forms only one object in solution, but is prone to precipitation once separated from MBP. Yet we have successfully obtained crystals of this fragment, which pave the way to solve its experimental 3D structure. Calcyanin is a difficult protein to work with, both experimentally and by bioinformatics. But we managed to model and characterize several of its fragments. From that, we inferred relevant information on calcyanin. More specifically, we highlighted a new family of domains, CoBaHMA, which presence on other protein architectures opens up new hints to understand its function and evolution. (1) Benzerara et al., 2022 14(3): evav026. doi :10.1093/gbe/evac026. (2) Gaschignard et al., En préparation
Cury, Jean. "Evolutionary genomics of conjugative elements and integrons." Thesis, Sorbonne Paris Cité, 2017. http://www.theses.fr/2017USPCB062/document.
Full textJanky, Rekin's. "Etude bioinformatique de l'évolution de la régulation transcriptionnelle chez les bactéries." Doctoral thesis, Universite Libre de Bruxelles, 2007. http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/210603.
Full textThe purpose of my thesis is to study the evolution of regulation within bacterial genomes by using a cross-genomic comparative approach. Nowadays, numerous genomes have been sequenced facilitating in silico analysis in order to detect groups of functionally related genes and to predict the mechanism of their relative regulation. In this project, we combined prediction of operons and regulons in order to reconstruct the transcriptional regulatory network for a bacterial genome. We have implemented three methods in order to predict operons from a bacterial genome and evaluated them on hundreds of annotated operons of Escherichia coli and Bacillus subtilis. It turns out that a simple distance-based threshold method gives good results with about 80% of accuracy. The principle of this method is to classify pairs of adjacent genes as “within operon” or “transcription unit border”, respectively, by using a threshold on their intergenic distance: two adjacent genes are predicted to be within an operon if their intergenic distance is smaller than 55bp. In the second part of my thesis, I evaluated the performances of a phylogenetic footprinting approach based on the detection of over-represented spaced motifs. This method is particularly suitable for (but not restricted to) Bacteria, since such motifs are typically bound by factors containing a Helix-Turn-Helix domain. We evaluated footprint discovery in 368 E.coli K12 genes with annotated sites, under 40 different combinations of parameters (taxonomical level, background model, organism-specific filtering, operon inference, significance threshold). Motifs are assessed both at the level of correctness and significance. The footprint discovery method proposed here shows excellent results with E. coli and can readily be extended to predict cis-acting regulatory signals and propose testable hypotheses in bacterial genomes for which nothing is known about regulation. Moreover, the predictive power of the strategy, and its capability to track the evolutionary divergence of cis-regulatory motifs was illustrated with the example of LexA auto-regulation, for which our predictions are remarkably consistent with the binding sites characterized in different taxonomical groups. A next challenge was to identify groups of co-regulated genes (regulons), by regrouping genes with similar motifs, in order to address the challenging domain of the evolution of transcriptional regulatory networks. We tested different metrics to detect putative pairs of co-regulated genes. The comparison between predicted and annotated co-regulation networks shows a high positive predictive value, since a good fraction of the predicted associations correspond to annotated co-regulations, and a low sensitivity, which may be due to the consequence of highly connected transcription factors (global regulator). A regulon-per-regulon analysis indeed shows that the sensitivity is very weak for these transcription factors, but can be quite good for specific transcription factors. The originality of this global strategy is to be able to infer a potential network from the sole analysis of genome sequences, and without any prior knowledge about the regulation in the considered organism.
Doctorat en Sciences
info:eu-repo/semantics/nonPublished
Bourquard, Thomas. "Exploitation des algorithmes génétiques pour la prédiction de structures protéine-protéine." Paris 11, 2009. http://www.theses.fr/2009PA112302.
Full textMost proteins fulfill their functions through the interaction with one or many partners as nucleic acids, other proteins…. Because most of these interactions are transitory, they are difficult to detect experimentally and obtaining the structure of the complex is generally not possible. Consequently, “in silico prediction” of the existence of these interactions and of the structure of the resulting complex has received a lot of attention in the last decade. However, proteins are very complex objects, and classical computing approaches have lead to computer-time consuming methods, whose accuracy is not sufficient for large scale exploration of the so-called “interactome” of different organisms. In this context development of high-throughput prediction methods for protein-protein docking is needed. We present here the implementation of a new method based on : Two types of formalisms : the Vornonoi and Laguerre tessellations, two simplified geometric models for coarse-grained modeling of complexes. This leads to computation time more reasonable than in atomic representation, the use and optimization of learning algorithms (genetic algorithms) to isolate the most relevant conformation between two two protein parteners, an evaluation method based on clustering of meta-attributes calculated at the interface to sort the best subset of candidate conformations
Estana, Alejandro. "Algorithms and comptutational tools for the study of Intrinsically Disordered Proteins." Thesis, Toulouse, INSA, 2020. http://www.theses.fr/2020ISAT0012.
Full textIntrinsically Disordered Proteins (IDPs) are involved in many biological processes. Their inherent plasticity facilitates very specialized tasks in cell regulation and signalling, and their malfunction is linked to severe pathologies. Understanding the functional roles of IDPs requires their structural characterization, which is extremely challenging, and needs a tight coupling of experimental and computational methods. In contrast to structured/globular proteins, IDPs cannot be represented by a single conformation, and their models must be based on ensembles of conformations representing a distribution of states that the protein adopts in solution. While purely random coil ensembles can be reliably constructed by available bioinformatics tools, these tools fail to reproduce the conformational equilibrium present in partially-structured regions.In this thesis, we propose several computational methods that, combined with experimental data, provide a better structural characterization of IDPs. These methods can be grouped in two main categories: methods to construct conformational ensemble models, and methods to simulate conformational transitions.Contributing to the first type of methods, we propose a new approach to generate realistic conformational ensembles that improves previously existing methods, being able to reproduce the partially-structured regions in IDPs.This method exploits structural information encoded in a database of three-residue fragments (tripeptides) extracted from high-resolution experimentally-solved protein structures.We have shown that conformational ensembles generated by our method reproduce accurately structural descriptors obtained from NMR and SAXS experiments for a benchmark set of nine IDPs. Also exploiting the tripeptide database, we have developed an algorithm to predict the propensity of some fragments inside IDPs to form secondary structure elements. This new method provides more accurate results than those of the most commonly-used predictors available on our benchmark set of well-characterized IDPs.Contributing to the second type of methods, we have developed an original approach to model the folding mechanism of secondary structural elements. The computation of conformational transitions is formulated as a discrete path search problem using the tripeptide database. To evaluate the approach, we have applied the strategy to two small synthetic polypeptides mimicking two common structural motifs in proteins. The folding mechanisms extracted are very similar to those obtained when using traditional, computationally expensive approaches. Finally, we have developed a more general method to compute transition paths between a (possibly large) set of conformations of an IDP. This method builds on a multi-tree variant of the TRRT algorithm, developed at LAAS-CNRS, and which provided good results for small and middle-sized biomolecules. In order to apply this method to IDPs, we have proposed a hybrid strategy for the parallelization of the algorithm, enabling an efficient execution in computer clusters.In addition to the aforementioned methodological work, I have been actively involved in multidisciplinary work, together with biophysicists and biologists,where I have applied these methods to the investigation of important biological systems, in particular the huntingtin protein, the causative agent of Huntington's disease.In conclusion, the work carried out during my PhD thesis has enabled a better understanding of the relationship between sequence and structural properties of IDPs, paving the way to novel applications. For example, this deeper understanding of sequence-structure relationships will enable us to anticipate structural perturbations exerted by sequence mutations, and subsequently, the rational design of IDPs with tailored properties for biotechnological applications
Autin, Ludovic. "Analyse des systèmes tenase et prothrombinase par bioinformatique structurale : prédiction de complexes macromoléculaires et proposition d'agents anti-coagulants." Paris 5, 2005. http://www.theses.fr/2005PA05P627.
Full textAnalysis of the tenase and prothrombinase systems by structural bioinformatic : macromolecular complexes prediction and proposition of new anticoagulant drug. The Tenase (F8a, F9a, F10) and Prothrombinase (F5a, F10a, PTH) complexes are essential in the blood coagulation. These complexes assembly are based on proteinprotein interactions which are not yet understood at the molecular levels. Thus, the bioinformatics lead us to a better comprehension of these interactions. And so, the most promising method is the " docking ", which permits to find the nearest interface between two molecules structures. This theoretical approach generates hundred of structural interfaces. Based on agreements with known experimental data, ten representative models of the tenase complex and one prothrombinase complex were selected. These structural models open the door of future experimentation helping clarify several ambiguous points and of future virtualscreening study in order to identify new lead able to bind pocket at proteinprotein interface
Traore, Seydou. "Computational approaches toward protein design." Thesis, Toulouse, INSA, 2014. http://www.theses.fr/2014ISAT0033/document.
Full textComputational Protein Design (CPD) is a very young research field which aims at providing predictive tools to complementprotein engineering. Indeed, in addition to the theoretical understanding of fundamental properties and function of proteins,protein engineering has important applications in a broad range of fields, including biomedical applications, biotechnology,nanobiotechnology and the design of green reagents. CPD seeks at accelerating the design of proteins with wanted propertiesby enabling the exploration of larger sequence space while limiting the financial and human costs at experimental level.To succeed this endeavor, CPD requires three ingredients to be appropriately conceived: 1) a realistic modeling of the designsystem; 2) an accurate definition of objective functions for the target biochemical function or physico-chemical property; 3)and finally an efficient optimization framework to handle large combinatorial sizes.In this thesis, we addressed CPD problems with a special focus on combinatorial optimization. In a first series of studies, weapplied for the first time the Cost Function Network optimization framework to solve CPD problems and found that incomparison to other existing methods, it brings several orders of magnitude speedup on a wide range of real CPD instancesthat include the stability design of proteins, protein-protein and protein-ligand complexes. A tailored criterion to define themutation space of residues was also introduced in order to constrain output sequences to those expected by natural evolutionthrough the integration of some structural properties of amino acids in the protein environment. The developed methods werefinally integrated into a CPD-dedicated software in order to facilitate its accessibility to the scientific community
Colcombet-Cazenave, Baptiste. "Structural and functional characterization of the protein PDZD7 as part of the Usher2 complex." Electronic Thesis or Diss., Sorbonne université, 2022. https://accesdistant.sorbonne-universite.fr/login?url=https://theses-intra.sorbonne-universite.fr/2022SORUS323.pdf.
Full textHearing relies on the capacity of specialized sensory hair cells in the cochlea to transduce sound-induced vibrations into electrical signals that are transmissible to the brain. Hair cells possess actin-filled stereocilia structured into staircase-shaped bundles deflected by sound-waves. Large protein complexes are found at the anchoring sites of extracellular links that interconnect stereocilia. Mutations of these proteins are responsible for hereditary sensory diseases, notably the Usher syndromes. The Usher syndrome of type 2 (Usher 2) is the most common form genetic cause of combined congenital deafness and progressive blindness. The Usher 2 protein complex, involved in the morphogenesis of the hair bundles, encompasses two large transmembrane proteins, usherin and the G protein-coupled receptor (GPCR) ADGRV1, with very large extracellular domains forming fibrous links between the stereocilia. These two proteins possess a cytoplasmic region in interaction with the scaffolding proteins whirlin and PDZD7, which in turn associate to actin-binding proteins. Usher proteins contain numerous protein-protein interaction domains necessary to the intricacy of the network, but the network’s assembly remains elusive, thus leaving the effect of mutations detected in patients to speculation. In this project, I studied the Usher 2 complex components from the molecules (in silico and in vitro) to the complex assembly at the cellular level. First, I deeply analyzed the sequences of the orphan domain family HHD, found in few neuronal proteins, using bioinformatics tools to predict functional surfaces involved in protein-protein interactions and the effect of human pathogenic mutations. Then, I characterized the determinants of interaction between the adhesion GPCR ADGRV1 and its PDZ-domain containing regulator PDZD7. I showed that the two N-terminal PDZ domains of PDZD7 are able to interact with ADGRV1 PDZ binding motif, with a higher affinity for the second PDZ. This interaction requires C-terminal extension of the PDZ domains which likely adopts a beta strand conformation in solution. I showed that two human pathological mutations of PDZD7 PDZ domains trigger a drastic decrease of affinity for ADGRV1, potentially disrupting the Usher2 complex. To further understand the activation mechanism of ADGRV1, I started its structural characterization in complex with its associated G protein by cryo electron microscopy. At the cellular level, I used high resolution STED microscopy to decipher the accurate localization of the Usher 2 complex and its anisotropic distribution between hair cell stereocilia rows. Finally, I initiated an ambitious cryo electron tomography project to solve the general organization of the Usher 2 complex in situ. To this end, I optimized the cochlea preparation for cryo-Correlative Light and Electron Microscopy Focused Ion Beam milling (cryo-CLEM FIB milling) and performed the first lift-out procedure on mouse dissected cochlea. Altogether, the obtained results will help to understand the physiopathology of mutations associated to the Usher syndrome of type 2
Nadaradjane, Aravindan. "Exploring the use of Deep Mutational Scanning and of Evolution for the Structural Prediction of Protein Complexes." Thesis, université Paris-Saclay, 2020. http://www.theses.fr/2020UPASS014.
Full textThe thesis project aimed at developing computational strategies to exploit the information generated by deep mutational scanning (DMS) technologies to predict the structures of protein assemblies. In that scope, I explored how to improve the agreement between the models simulated by molecular docking and experimental constraints. From the literature, two reference complexes whose structures have been solved experimentally and for which DMS data were published could be used for the methodological development: the parD3-parE3 and dockerin-cohesin complexes. For each of the many mutants generated by DMS, an experimental score quantifying the affinity of the complex could be extracted from the available data. For the simulations, a number of protocols based on the Rosetta software were tested and optimized to predict the effect of mutations on interface stability. A compromise was found between efficiency and precision, allowing for a fair estimation of the effect of mutations on native complex structures. The agreement between the predicted and the experimental data was quantified using two different metrics, either the correlation between the predicted and experimental binding scores or the area under the ROC (Receiver Operating Characteristic) curve, defining how efficiently the predictor could sort out the most impacting mutations. Applied to a set of 1000 decoys of complexes generated by docking, both metrics were assessed for their ability to discriminate correct from wrong models. For both reference systems, the second metrics based on ROC curves was found most useful. This methodology was further applied to an antibody-antigen complex which was studied by DMS in the group of B. Maillère. My PhD work was also dedicated to the processing of the raw data from DMS experiments which were generated by our collaborators, O. Pereira-Ramos and L. Martin, in order to design a high affinity peptide for the protein Asf1 and to screen interaction surfaces between Asf1 and its binding partners. Last, throughout my PhD I had the opportunity to participate in all targets submitted to the docking community by the organizers of CAPRI, an international challenge for the assessment of methods for the structural prediction of protein interactions. The manuscript details all the strategies which were set up to tackle these challenges for which our team eventually ranked first by generating the highest number of both correct and precise models
Candat, Adrien. "Analyse de la localisation subcellulaire des protéines LEA (late embryogenesis abundant) chez Arabidopsis thaliana par des approches de bioinformatique et de biologie cellulaire." Angers, 2012. http://www.theses.fr/2012ANGE0048.
Full textAnhydrobiosis is the ability to survive severe desiccation and to resume normal metabolism upon a return to favorable conditions of water availability. It is a phenomenon based on mutiple factors, including the accumulation of stress proteins such as LEA (Late Embryogenesis Abundant) proteins. These proteins, which in the native state are generally very hydrophilic and disordered, are clustered into several families based on their primary sequence. In the model plant Arabidopsis thaliana, 51 genas encoding LEA proteins have been previously identified. The purpose of this work was to characterize the subcellular localization of these LEA proteins in order to better understand their, as yet, enigmatic functions. Bioinformatic analyses and experimental approaches, based on transient expression of fluorescent fusion proteins in Arabidopsis protoplasts, or seedlings, were used. The experimental data highlight the limits of in silico predictions for analysis of subcellular localization, and will help to improve prediction algorithms. An original method to accurately identify the cleavage sites of targeting peptides for organellar proteins has also been developed. The combination of experimentally determined subcellular location and identity of the mature LEA proteins is essential for the accurate design of corresponding recombinant proteins. Finally, examination of the relationship between the classification of LEA proteins, their gene expression, and their subcellular localization, enabled the development of novel hypotheses with respect to the putative functions of this important group of proteins
Moine-Franel, Alexandra. "Cartographie des poches aux interfaces protéine-protéine et identification de nouvelles cibles thérapeutiques potentielles." Electronic Thesis or Diss., Sorbonne université, 2023. https://accesdistant.sorbonne-universite.fr/login?url=https://theses-intra.sorbonne-universite.fr/2023SORUS634.pdf.
Full textProtein-protein interactions (PPIs) constitute a significant source of potential therapeutic targets because they play a crucial role in numerous and diverse biological processes, including the development of pathologies. While PPIs appear as promising therapeutic targets, they are more challenging to study than conventional therapeutic targets. Indeed, known PPIs are characterized by specific structural motifs that limit their ‘druggability’, meaning their ability to bind to and be modulated by a small drug molecule. However, the growing identification of small molecules modulating various PPIs demonstrates that, with an appropriate methodology, they can represent a class of novel and innovative therapeutic targets. The objective is, therefore, to develop an in silico protocol to aid in identifying new therapeutic targets involving PPIs by rationalizing the key elements that determine the ‘druggability’ of the interaction
Albou, Laurent-Philippe. "Analyse intégrative des données structurales et reconnaissance de forme : application à la régulation de la transcription eucaryote." Strasbourg, 2010. https://publication-theses.unistra.fr/public/theses_doctorat/2010/ALBOU_Laurent-Philippe_2010.pdf.
Full textIn 5 years, international projects of Structural Biology and Structural Genomics have doubled the number of available molecular structures in the Protein Data Bank. During this thesis, I have developped Structural Bioinformatic approaches to perform the integrated analysis of structural data, to better describe the molecular mechanisms of interactions. We have shown that, on average, 44% of protein surfaces are involved in interactions with molecules other than solvants and ions. If 86% of protein surfaces can be transiently hydrated, only 15% can be specifically hydrated. By differentiating every type of binding sites (protein, DNA, RNA, ligand…) of each protein, we have shown the existence of overlaps between these regions. This observation has led us to define two major families of binding sites : specific sites, which can only bind one type of molecule, and polyvalent sites, which can bind at least two different types of molecule. The specific binding sites differ greatly from polyvalent ones, in particular in terms of hydrophobicity. Specific binding sites may indicate stronger or permanent interactions. The fast and systematic analysis of molecular surfaces has also required the development of advanced geometrical approaches, based on alpha shapes, to define contiguous regions and local curvatures. The screening of these contiguous regions, like a blast but for local 3D regions, open the way to numerous biological and pharmaceutical applications
Pages, Guillaume. "Développements algorithmiques pour l'analyse et la prédiction de la structure des protéines." Thesis, Université Grenoble Alpes (ComUE), 2019. http://www.theses.fr/2019GREAM036.
Full textProteins are ubiquitous for virtually all biological processes. Identifying their role helps to understand and potentially control these processes. However, even though protein sequence determination is now a routine procedure, it is often very difficult to use this information to extract relevant functional knowledge about system under study. Indeed, the function of a protein relies on a combination of its chemical and mechanical properties, which are defined by its structure. Thus, understanding, analysis and prediction of protein structure are the key challenges in molecular biology.Prediction and analysis of individual protein folds is the central topic of this thesis. However, many proteins are organized in higher-level assemblies, which are symmetric in most of the cases, and also some proteins contain internal repetitions.In many cases, designing a fold with repetitions or designing a symmetric protein assembly is the simplest way for evolution to achieve a specific function. This is because the number of combinatorial possibilities in the interactions of designed folds reduces exponentially in the symmetric cases. This motivated us to develop specific methods for symmetric protein assemblies and also for individual proteins with internal repeats. Another motivation behind this thesis was to explore and advance the emerging deep neural network field in application to atomistic 3-dimensional (3D) data.This thesis can be logically split into two parts. In the first part, we propose algorithms to analyse structures of protein assemblies, and more specifically putative structural symmetries.We start with a definition of a symmetry measure based on 3D Euclidean distance, and describe an algorithm to efficiently compute this measure, and to determine the axes of symmetry of protein assemblies. This algorithm is able to deal with all point groups, which include cyclic, dihedral, tetrahedral, octahedral and icosahedral symmetries, thanks to a robust heuristic that perceives correspondence between asymmetric subunits. We then extend the boundaries of the problem, and propose a method applicable to the atomistic structures without atom correspondence, internal symmetries, and repetitions in raw density maps. We tackle this problem using a deep neural network (DNN), and we propose a method that predicts the symmetry order and a 3D symmetry axis.Then, we extend the DNN architecture to recognise folding quality of 3D protein models. We trained the DNN using as input the local geometry around each residue in a protein model represented as a density map, and we predicted the CAD-scores of these residues. The DNN was specifically conceived to be invariant with respect to the orientation of the input model. We also designed some parts of the network to automatically recognise atom properties and robustly select features. Finally, we provide an analysis of the features learned by the DNN. We show that our architecture correctly learns atomic, amino acid, and also higher-level molecular descriptors. Some of them are rather complex, but well understood from the biophysical point of view. These include atom partial charges, atom chemical elements, properties of amino acids, protein secondary structure and atom solvent exposure. We also demonstrate that our network learns novel structural features.This study introduces novel tools for structural biology. Some of them are already used in the community, for example, by the PDBe database and CASP assessors. It also demonstrates the power of deep learning in the representation of protein structure and shows applicability of DNNs to computational tasks that involve 3D data
Friedrich, Anne. "De la mutation structurale aux phénotypes des pathologies animales : vers une approche intégrative des mutations et de leurs conséquences." Université Louis Pasteur (Strasbourg) (1971-2008), 2007. https://publication-theses.unistra.fr/public/theses_doctorat/2007/FRIEDRICH_Anne_2007.pdf.
Full textThis thesis work focuses on high throughput applications of bioinformatics methodologies to study genotype/phenotype correlations in the context of the MS2PH project (“from Structural Mutation to Human Pathologies Phenotypes”). In an initial step, we concentrated on the characterization of mutations in an evolutionary context and we have developed an optimal strategy for the construction of multiple alignments of protein sequences dedicated to structural analysis. Next, the characterization of the mutations in a functional and structural context was addressed by the development of MAGOS, a Web server which allows the coupling of the sequence, structural, functional and evolutionary information related to a given protein of interest. Finally, we exploited the integrative capabilities of MAGOS, as well as the power of the Decrypthon computation grid (www. Decrypthon. Fr) to develop MS2PH-db, a database dedicated to proteins involved in human monogenic diseases, which also integrates clinical data (mutations and associated phenotypes). In the final section, the basic associative rules between sequence, mutation, structural impact and clinical phenotypes are discussed, illustrated by the analysis of two well-studied proteins for which the genotype/phenotype correlations are partially elucidated. These studies open up the way to the development of a system dedicated to the prediction of the link between mutations and clinical phenotypes
Beinsteiner, Brice. "Origine et évolution des récepteurs nucléaires et étude structurale du premier stéroïdien, ERR." Thesis, Strasbourg, 2018. http://www.theses.fr/2018STRAJ099.
Full textNuclear receptors (NRs) are transcription factors which bind to specific DNA sequences and activate gene transcription in response to the binding of specific ligands. Among all of the RNs involved in the etiology of cancers, ERR estrogen receptors play an important role in breast, ovarian, colon, endometrial and prostate cancers. This NR is said to be orphan because it does not have a natural ligand known to date. Using an integrative structural biology approach combining cryo-electron microscopy, bioinformatics and evolution, my PhD work focused on the structural study of ERR and the origin and evolution of RNs. In this context, three informatic tools have been developed. The results obtained allowed, on the one hand, the revision of fundamental knowledge on the origin of nuclear receptors and their evolution. On the other hand, structural study of ERR allow us to acquire new data on topology of steroid nuclear receptors fixed on an element of ERRE / ERE response as well as on the allosteric mechanism of the binding of the coactivator PGC-1α on the dimer of ERR. The resolution of the complex at the atomic scale by cryo-electron microscopy will open the way towards the design of new therapeutic molecules
Flutre, Timothée. "L'annotation des éléments transposables par la compréhension de leur diversification." Phd thesis, Université Paris-Diderot - Paris VII, 2010. http://tel.archives-ouvertes.fr/tel-00560242.
Full textFerrario, Maria Giovanna. "On the recognition of ecdysteroids by the ecdysone receptor : a computational study." Strasbourg, 2010. https://publication-theses.unistra.fr/restreint/theses_doctorat/2010/FERRARIO_Maria_Giovanna_2010.pdf.
Full textGelly, Jean-Christophe. "Système d'information et outils de prédiction structurale spécifiques de classes de protéines : Base de données KNOTTIN et matrices de substitution EvDTree dépendantes de la structure." Montpellier 2, 2004. http://www.theses.fr/2004MON20026.
Full textTrinh, Minh Hieu. "Modélisation de l'assemblage de protéines multi-domaines avec des contraintes expérimentales de microscopie à force atomique." Thesis, Montpellier 2, 2010. http://www.theses.fr/2010MON20076/document.
Full textA major challenge in the field of structural biology is to obtain high-resolution information on the major biological macromolecules. Because of their size and their flexibility, the traditional techniques of structural biology are often powerless. One of the promising techniques is atomic force microscopy (AFM). Unlike optical microscopy, AFM uses a mechanical probe of very small size (<10 nm) to obtain topographical information on isolated biological material deposited on ultra flat surfaces. The aim of the thesis was to develop tools to enable the modeling of large macromolecules at the atomic level while incorporating topological constraints obtained by AFM imaging. Using high resolution AFM height images, a protocol for assembling protein domains has been developed. It uses an exhaustive search in real three-dimensional space of all possible orientations of the macromolecule's domains respecting the boundaries imposed by the AFM topographical image. A set of distance constraints between each of the domains allows an initial screening of candidate models. A final ranking is assigned to each model according to a score called EFactor, estimator of the similarity between the experimental topography and the model. The protocol was validated on model systems that are antibodies. It was also used to reconstruct a virus particle (tobacco mosaic virus) and assemble the tetrameric structure of the membrane protein aquaporin Z
Friedrich, Anne Poch Olivier. "De la mutation structurale aux phénotypes des pathologies animales vers une approche intégrative des mutations et de leurs conséquences /." Strasbourg : Université Louis Pasteur, 2007. http://eprints-scd-ulp.u-strasbg.fr:8080/898/01/FRIEDRICH_Anne_2007.pdf.
Full textDesaphy, Jérémy. "L'analyse structurale de complexes protéine/ligand et ses applications en chémogénomique." Phd thesis, Université de Strasbourg, 2013. http://tel.archives-ouvertes.fr/tel-00997394.
Full textGaliez, Clovis. "Fragments structuraux : comparaison, prédictibilité à partir de la séquence et application à l'identification de protéines de virus." Thesis, Rennes 1, 2015. http://www.theses.fr/2015REN1S124/document.
Full textThis thesis investigates the local characterization of protein families at both structural and sequential level. We introduce contact fragments (CF) as parts of protein structure that conciliate spatial locality together with sequential neighborhood. We show that the predictability of CF from the sequence is better than that of contiguous fragments and of structurally distant pairs of fragments. In order to structurally compare CF, we introduce ASD, a novel alignment-free dissimilarity measure that respects triangular inequality while being tolerant to sequence shifts and indels. We show that ASD outperforms classical scores for fragment comparison on practical experiments such that unsupervised classification and structural mining. Ultimately, by integrating the identification of CF from the sequence into a statistical machine learning framework, we developed VIRALpro, a tool that enables the detection of sequences of viral structural proteins
Lindenbaum, Pierre. "Roxan, une nouvelle proteine cellulaire interagissant avec la proteine non-structurale nsp3 du rotavirus : clonelt* : un programme en ligne trouvant des strategies de clonage (doctorat : microbiologie)." Paris 11, 2000. http://www.theses.fr/2000PA114811.
Full textMahmoudi, Ikram. "Structural and evolutionary analysis of protein-RNA interfaces and prediction perspectives." Electronic Thesis or Diss., université Paris-Saclay, 2024. http://www.theses.fr/2024UPASQ024.
Full textProtein-RNA interactions are crucial in numerous cellular pathways and pathologies. Knowledge of their 3D structures is critical for understanding their functions, yet their experimental determination remains challenging. The scarcity of structural data and the inherent flexibility of these complexes have hindered the advancement of protein-RNA interface structural prediction. At the same time, tremendous progress has been made recently for protein-protein interaction prediction thanks to methods leveraging evolutionary information and deep learning.My thesis focused on a detailed evolutionary analysis of protein-RNA interface structures. I first identified 2,022 pairs of structurally homologous interfaces. I explored the conservation of interface contacts among these pairs, discovering a high conservation rate for distance-based and apolar contacts, even in distant homologs. Hydrogen bonds, salt bridges, and π-stacking interactions displayed higher versatility. I investigated mechanisms compensating for non-conserved interactions. I contributed to developing a web interface allowing the community to explore evolutionary structural insights in our datasets. I also participated in a collaborative project with biologists to study a specific protein-RNA interface.Then, I investigated how to incorporate evolutionary signals into protein-RNA structural modeling methods using machine learning models, including logistic regression and CatBoost classifiers. I assessed these models' ability to learn how to transfer contacts from remote interologs and generalize across datasets while mitigating overfitting. Lastly, I explored developing functions based on contact propensities to score protein-RNA docking poses. These efforts constitute a step towards improving protein-RNA structure prediction
Devillé, Julie. "Etude structurale des cassures d'hélices et son application à la modélisation des récepteurs couplés aux protéines G (RCPG)." Phd thesis, Université d'Angers, 2007. http://tel.archives-ouvertes.fr/tel-00346950.
Full textVelusamy, Mahesh. "New computational approaches for investigating the impact of mutations on the transglucosylation activity of sucrose phosphorylase enzyme." Thesis, La Réunion, 2018. http://www.theses.fr/2018LARE0045.
Full textIn this thesis, we explore the usage of computational approaches for understanding the link between mutations and changes in protein activity. Our study model is a bacterial sucrose phosphorylase enzyme from Bifidobacterium adolescentis (BaSP). This glycosyl hydrolase from family 13 (GH13) has been a focus in the industry due to its ability to synthesize original disaccharides and glycoconjugates. In fact, its activity is to transfer a glucose moiety from a donor sucrose to an acceptor which can be a monosaccharide or a hydroxylated aglycone. The enzymatic reaction proceeds by a double displacement with retention of configuration mechanism whereby a covalent glucosyl-enzyme intermediate is formed. However, it is at stake to control the regioselectivity of this transfer for it to be applicable at industrial level. This thesis aimed at providing a rational explanation for the observed impact of mutations on the regioselectivity of BaSP in view of controlling the synthesis of rare pre-biotic disaccharides like kojibiose and nigerose. We hypothesized that the preferred orientations of the acceptor determines the regioselectivity of the enzyme. In that respect, we used computational approaches to investigate the impact of mutations on the binding of the acceptor to the glucosyl-enzyme intermediate. The methodology used in this work opens the perspective of using computational approaches for engineering the regioselectivity of of glycosyl hydrolases with similar mechanism
Schwarz, Benjamin. "Application de la théorie des formes alpha pour la caractérisation de la surface et des poches de macromolécules biologiques." Strasbourg, 2009. http://www.theses.fr/2009STRA6196.
Full textOur study is concerned with structural bioinfomatics (aka computational biology), more specifically, we borrow models from the alpha-shape theory to represent and study molecules. Roughly, our aim is to provide new theoretical and practical tools to ease the study of structure-function relationship in biological molecules. We are more specifically interested in characterising the usual locations of a possible interaction at the surface of such molecules. In this context we propose a novel model, the dual surface, that constitutes a manifold polyhedral surface encoding the Accessible surface. This construction eases the the construction of continuous surface tracks at the surface of a molecule, and therefore allows notably, the construction of molecular surface patches. We adapted this model mainly to address three distinct problems : (a) the proposal of a novel index to describe the molecular surface landscape in terms of knobs and clefts, (b) the definition of surface descriptors that can be used to study interacting patches on a protein surface, (c) the detection and characterisation of cavities, pockets, clefts and crevices at the surface of macromolecules. Two software tools were developped based on these works and are now freely accessible to the scientific community : LC and Pck, respectively dedicated to the description of the molecular surface topography, and to the detection and characterisation of pockets in molecular structures
Legendre, Audrey. "Prédiction de structures secondaires d’ARN et de complexes d’ARN avec pseudonoeuds - Approches basées sur la programmation mathématique multi-objectif." Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLE031.
Full textIn this thesis, we propose new algorithms and tools to predict RNA and RNA complex secondary structures, including particular RNA motifs, difficult to predict, like pseudoknots. RNA structure prediction stays a difficult task, and the numerous existing tools don't always give good predictions.In order to predict structures that are as close as possible to the real ones, we propose to develop algorithms that: i) predict the k-best structures; ii) combine several models of prediction to take advantage of each; iii) are able to take into account user constraints and structural data like SHAPE.We developed three tools: BiokoP for predicting RNA secondary structures and RCPred and C-RCPred for predicting RNA complex secondary structures.The tool BiokoP proposes several optimal and sub-optimal structures thanks to the combination oftwo prediction models, the energy model MFE and the probabilistic model MEA. This combination isdone with multi-objective mathematical programming, where each model is associated to an objective function. To this end, we developed a generic algorithm returning the k-best Pareto curves of a bi-objective integer linear program.The tool RCPred, based on the MFE model, proposes several sub-optimal structures. It takes advantage of the numerous existing tools for RNA secondary structure prediction and for RNA-RNA interaction prediction, by taking as input predicted secondary structures and RNA-RNA interactions. The goal of RCPred is to find the best combination among these inputs.The tool C-RCPred is a new version of RCPred, taking into account user constraints and structural data(SHAPE, PARS, DMS). C-RCPred is based on a multi-objective algorithm, where the different objectives are the MFE model, the fulfillment of the user constraints and the concordance with the structural data
Machat, Mohamed. "Computational geometry for the determination of biomolecular structures." Thesis, Paris 6, 2017. http://www.theses.fr/2017PA066359/document.
Full textStructural biology has allowed us expand our knowledge of living organisms. It is defined as the investigation of the structure and function of biological systems at the molecular level. Studying a biomolecule's structure offers insight into its geometry, as angles and distances between the biomolecule's atoms are measured in order to determine the biomolecular structure. The values of these geometrical parameters may be obtained from biophysical techniques, such as X-ray crystallography or nuclear magnetic resonance (NMR) spectroscopy. One of the most used methods to calculate protein structures from geometric restraints is simulated annealing. This method does not guarantee an exhaustive sampling of protein conformational space, which is a shortcoming as one protein may adopt multiple functional conformations, and it is important to determine them exhaustively. In this PhD project, the efficiency of a new method - derived from operations research and computational geometry - is studied in order to answer this question: How does this method explore the conformational spaces of small proteins? This method - implemented within the iBPprot software framework - treats protein structure determination as a distance geometry problem, which the interval branch-and-prune algorithm tries to solve by the full exploration of its solutions space. The results obtained by iBPprot on a set of test proteins, with sizes ranging from 24 to 120 residues and with known structures, are analyzed here. Using short-range exact distance restraints, it was possible to rebuild the structure of all protein targets, and for many of them it was possible to exhaustively explore their conformational spaces. In practice, it is not always possible to obtain exact distance restraints from experiments. Therefore, this method was then tested with interval data restraints. In these cases, iBPprot permitted the sampling of the positions of more than 70% of the atoms constituting the protein backbone for most of the targets. Furthermore, conformations whose r.m.s. deviations closer than 6 Angstrom to the target ones were obtained during the conformational space exploration. The quality of the generated structures was satisfactory with respect to Ramachandran plots, but needs improvement because of the presence of steric clashes in some conformers. The runtime for most performed calculations was competitive with existing structure determination method
Chèneby, Jeanne. "Etudes des éléments cis-régulateurs : identification et caractérisation." Thesis, Aix-Marseille, 2019. http://www.theses.fr/2019AIXM0520.
Full textThe regulation of gene transcription is largely based on the existence of non-coding DNA sequences in the genome. These DNA sequences, called "cis-regulatory elements", have the particularity of recruiting many proteins capable of regulating the level of gene transcription. The direct or indirect binding of these regulatory proteins to cis-regulatory elements allows the regulation of genes in space and time. The massive accumulation of sequencing data in public databases allows the integration of many experiments that capture the interactions between regulatory proteins and DNA through bioinformatics. The purpose of my PhD was to annotate and process in a uniform way the raw data from sequencing experiments whose objective is to identify the binding regions of regulatory proteins for humans and then for Arabidopsis Thaliana. We processed data from ChIP-seq, ChIP-exo and DAP-seq to develop several catalogues of regulatory regions. All this data is available within the ReMap project. To carry out these analyses we have developed reproducible, scalable and portable workflows on different architectures. These data were also used to identify the binding sites recognized by the transcription factors and to consolidate the JASPAR database. Finally, this catalogue was used in the development of a new method to differentiate between direct and indirect protein binding events in ChIP-seq results
Schweke, Hugo. "Développement d’une méthode in silico pour caractériser le potentiel d’interaction des surfaces protéiques dans un environnement encombré." Thesis, Université Paris-Saclay (ComUE), 2018. http://www.theses.fr/2018SACLS554.
Full textIn the crowded cell, proteins interact with their functional partners, but also with a large number of non-functional partners that compete with the functional ones. The goal of this thesis is to characterize the physical properties and the evolution of protein surfaces in order to understand how selection pressure exerts on proteins, shaping their interactions and regulating this severe competition.To do this I developed a framework based on docking calculations to characterize the propensity of protein surfaces to interact with other proteins. Molecular cartography enables the visualization and the comparison of surface properties of proteins. I implemented a new theoretical framework based on the representation of interaction energy landscapes by 2-D energy maps. These maps reflect in a synthetic manner the propensity of the surface of proteins to interact with other proteins. These maps are useful from a practical point view for determining the regions of protein’s surface that are more prone to interact with other proteins. Our new theoretical framework enabled to show that the surface of proteins harbor regions with different levels of propensity to interact with other proteins (hot regions, intermediate and cold regions to favorable, intermediate and unfavorable regions respectively).A large part of this thesis work consisted in characterizing the physico-chemical properties and the evolution of these regions. The other part of this thesis work consisted in applying this methodology on several study systems: homomeric complexes, cytosolic proteins from S. cerevisiae, families of interologs. This work opens the way to numerous practical applications in structural bioinformatics, such as binding site prediction, functional annotation and the design of new interactions.To conclude, the strategy implemented in this work enable the exploration of the propensity of a protein to interact with hundred of protein partners. It thus enables the investigation of the behavior of a protein in a crowded environment. This application goes beyond the classical use of protein docking as a, because our strategy provides a systemic point of view of protein interactions at an atomic resolution
Bricout, Alexandre. "Mise en évidence d’une forte diversité structurale de lipopeptides chez P. syringae, un complexe bactérien aux activités antifongiques prometteuses." Electronic Thesis or Diss., Université de Lille (2018-2021), 2020. http://www.theses.fr/2020LILUR011.
Full textThe use of microorganisms or natural substances of microbial origin is one of the identified alternatives to partially or totally replace conventional pesticides. In this context, the aim of this thesis was to evaluate the biocontrol potential of strains belonging to the P. syringae complex. First, the lipopeptides produced by strains of this bacterial complex have been studied because these molecules are known for their antimicrobial activities. Then, the antifungal activity of these bacteria was analysed. To reach this goal, a collection of 709 strains, representative of the phylogenetic diversity of the P. syringae complex, was explored. Through a strategy involving complementary approaches of mass spectrometry and bioinformatics, it has been possible to reveal a huge lipopeptide structural diversity: in total, 61 lipopeptides, including 38 new, distributed into the 5 families described in the P. syringae complex (syringafactin, syringomycin, corpeptin, syringopeptins 22 and 25) have been identified. Lipopeptides producing strains, which represent 81.1% of the collection studied, belong to 8 of the 13 phylogroups referenced in the P. syringae complex. Concerning their activities, 22.3% of the strains have shown an antifungal activity in vitro. Lipopeptides, which are produced by 97,3% of the antifungal strains and are also found in crude and ultra-filtered cell free supernatants, are certainly responsible for these activities. Finally, two strains have shown, in planta, an interesting potential for the biocontrol of Septoria tritici blotch of wheat caused by the fungus Zymoseptoria tritici. Their crude and ultra-filtered cell free supernatants have shown different wheat protection levels up to 62% compared to the infection control
Gianfrotta, Coline. "Modélisation, analyse et classification de motifs structuraux d'ARN à partir de leur contexte, par des méthodes d'algorithmique de graphes." Electronic Thesis or Diss., université Paris-Saclay, 2022. http://www.theses.fr/2022UPASG056.
Full textIn this thesis, we study the structural context of RNA structural motifs in order to make progress in their prediction. Indeed, some RNA motifs, which are substructures appearing recurrently in RNA structures, remain difficult to predict, because of the presence of non-canonical interactions in these motifs, and because of the distance on the primary sequence between the different parts of these motifs. We therefore model the topological structural context of these motifs by graphs, and compare the contexts of the different occurrences using several graph algorithms. We then classify the motif occurrences according to their topological context similarities and according to their 3D context similarities, using an overlapping clustering algorithm.First, we show on a dataset of three structural motifs that the observed similarities between the topological contexts are consistent with the similarities between the 3D contexts. This indicates that the topological context may be sufficient to determine the 3D context for these three motifs.In a second step, we study several classifications of occurrences of the A-minor motif, according to 3D context similarities. We observe that 3D context similarities exist between non-homologous occurrences, which could be a sign of an evolutionary convergence phenomenon. Moreover, we observe that some parts of the 3D context seem to be better conserved than others between non-homologous occurrences.In a third step, we study the predictive ability of the common topological context of A-minor motif occurrences, sharing similar 3D contexts, as well as the predictive ability of a sequence signal on these same occurrences. To this end, we study the occurrence of this topology and sequence in RNA structures in the absence of A-minor motifs. We conclude that the topology and the sequence represent a good signal for the majority of the studied classes
Douguet, Dominique. "Etude des interactions protéine-protéine et protéine-ligand par bio- et chimie-informatique structurale : Identification de petites molécules bio-actives." Habilitation à diriger des recherches, Université de Nice Sophia-Antipolis, 2007. http://tel.archives-ouvertes.fr/tel-00320089.
Full textLa modélisation par homologie permet d'obtenir un modèle tridimensionnel d'une protéine lorsque sa structure n'a pas été déterminée expérimentalement. Ma contribution dans ce domaine fut la réalisation du serveur @TOME avec le soutien de la GENOPOLE Languedoc-Roussillon (accessible à l'adresse http://bioserver.cbs.cnrs.fr). Ce serveur était le premier de ce type à avoir été développé en France. Le serveur @TOME rassemble et traite d'une manière automatique toutes les étapes nécessaires à la construction d'un modèle 3D d'une protéine. Cela inclut la reconnaissance du repliement, la construction des modèles protéiques et leur évaluation. Les résultats du CASP5 en 2005 (session internationale d'évaluation des méthodes de prédiction de la structure des protéines ; http://predictioncenter.llnl.gov/) ont montré que notre serveur utilisé en mode automatique propose des modèles très proches de la structure expérimentale lorsque l'identité de séquence avec la structure support est supérieure à 30%. Le serveur a été classé 26ième sur 187 groupes inscrits.
Dans un second temps, mes recherches m'ont permis de réaliser une base de données de complexes protéiques co-cristallisés, base fondatrice du projet DOCKGROUND. Ce projet de grande envergure, soutenu par le NIH depuis 2005, vise à établir un système intégré et dynamique de bases de données dédié à l'étude et à la prédiction des interactions entre protéines et permettre ainsi d'améliorer nos connaissances des interactions et de développer des outils de prédiction plus fiables. Ce travail a été effectué au sein de l'équipe du Pr. Ilya Vakser à l'Université de Stony Brook, NY, USA. Dans la réalisation de cette première base de données, un ensemble de programmes collectent, classent et annotent les complexes protéiques qui ont été co-cristallisés (données sur la séquence, la fonction, le repliement 3D, les particularités telles qu'une fixation à de l'ADN, ...). Ensuite, j'ai mis en œuvre une sélection dynamique des représentants des complexes contenus dans cette base. Les représentants sont essentiels pour éviter une surreprésentation de certaines familles de protéines. Cette base de donnée est accessible par Internet et est régulièrement mise à jour (http://dockground.bioinformatics.ku.edu). Le projet DOCKGROUND va être poursuivi par la réalisation de 3 autres bases de données qui s'ancreront sur la présente appelée ‘Bound-Bound'.
L'objectif principal de mes travaux est d'identifier de nouveaux composés bio-actifs afin de comprendre le fonctionnement de leur cible dans un contexte biologique. Les méthodes que j'utilise se basent sur la chémoinformatique, le criblage virtuel et le de novo ‘drug design'. Dans le cadre de ce dernier, j'ai mis au point un programme propriétaire LEA3D (‘Ligand by Evolutionary Algorithm' 3D). Le programme génère des petites molécules à partir de la combinaison de fragments moléculaires issus de drogues et de molécules ‘bio' (substrats ou produits de réactions enzymatiques). Le criblage virtuel basé sur la structure protéique et le de novo ‘drug design' par LEA3D, ont été appliqués avec succès à la thymidine monophosphate kinase (TMPK) de Mycobacterium tuberculosis dans le cadre d'une collaboration avec une équipe de chimistes et de biologistes de l'Institut Pasteur. De nouvelles familles d'inhibiteurs ont été identifiées dont un inhibiteur synthétique trois fois plus affin que le substrat naturel. Plusieurs publications et une demande de brevet couvrent les résultats de ces recherches. Dans la continuité de ces travaux, je m'intéresse maintenant, plus particulièrement, à développer des stratégies de criblages de fragments (molécules de petit poids moléculaire). Il a été montré que de petites chimiothèques contenant des petites molécules polaires sont plus efficaces pour identifier des touches. Ce travail doit être réalisé conjointement avec des criblages structuraux expérimentaux comme la RMN ou la diffraction des rayons X. Ces derniers se posent comme une alternative aux tests in vitro avec pour avantage de donner une information détaillée, au niveau atomique, des interactions entre le ligand et sa cible. S'ensuit une étape d'optimisation/maturation des touches en ligands plus élaborés et plus affins par l'utilisation d'outils de chémoinformatique.
Bricout, Alexandre. "Mise en évidence d’une forte diversité structurale de lipopeptides chez P. syringae, un complexe bactérien aux activités antifongiques prometteuses." Thesis, Lille 1, 2020. http://www.theses.fr/2020LIL1R011.
Full textThe use of microorganisms or natural substances of microbial origin is one of the identified alternatives to partially or totally replace conventional pesticides. In this context, the aim of this thesis was to evaluate the biocontrol potential of strains belonging to the P. syringae complex. First, the lipopeptides produced by strains of this bacterial complex have been studied because these molecules are known for their antimicrobial activities. Then, the antifungal activity of these bacteria was analysed. To reach this goal, a collection of 709 strains, representative of the phylogenetic diversity of the P. syringae complex, was explored. Through a strategy involving complementary approaches of mass spectrometry and bioinformatics, it has been possible to reveal a huge lipopeptide structural diversity: in total, 61 lipopeptides, including 38 new, distributed into the 5 families described in the P. syringae complex (syringafactin, syringomycin, corpeptin, syringopeptins 22 and 25) have been identified. Lipopeptides producing strains, which represent 81.1% of the collection studied, belong to 8 of the 13 phylogroups referenced in the P. syringae complex. Concerning their activities, 22.3% of the strains have shown an antifungal activity in vitro. Lipopeptides, which are produced by 97,3% of the antifungal strains and are also found in crude and ultra-filtered cell free supernatants, are certainly responsible for these activities. Finally, two strains have shown, in planta, an interesting potential for the biocontrol of Septoria tritici blotch of wheat caused by the fungus Zymoseptoria tritici. Their crude and ultra-filtered cell free supernatants have shown different wheat protection levels up to 62% compared to the infection control
Moniot, Antoine. "Modélisation 3D de complexes ARN-protéine par assemblage combinatoire de fragments structuraux." Electronic Thesis or Diss., Université de Lorraine, 2022. http://www.theses.fr/2022LORR0339.
Full textThe characterization of RNA-protein complexes at the atomic scale allows us to better understand the biological functions of these complexes, and to define therapeutic targets to regulate the biological phenomena in which they participate. The aim of this thesis is to develop tools to predict the structure of a protein-RNA complex when a 3D structure of the protein is known as well as the secondary structure of the interacting RNA part. We focus on the case where RNA is mainly in single-stranded form (unpaired nucleotides), raising the difficulty of its flexibility.A docking method developed in the CAPSID team is based on the use of structural fragments of single-stranded RNA. The work of this thesis builds on this method to perform docking of RNA secondary structures. We first evaluated the contribution of a loop closure constraint for docking the single-stranded loop of a hairpin structure, and then addressed the docking of the double-stranded elements of these structures, paving the way for the assembly of the entire complex.This fragment-based docking method is dependent on the use of structural fragment libraries. These libraries are composed of prototypes that represent the conformational landscape experimentally observed in protein-bound RNA structures. A large part of the thesis work consisted in the creation and optimization of such fragment libraries.We created the ProtNAff tool that allows to extract subsets of structures from the PDB and to create libraries of nucleic acid fragments, following complex combinations of criteria. It has been designed to exceed our needs, so that it can be adopted by the community for the treatment of various problems.We have developed a new approach for inferring prototypes of a set of conformations. The set of prototypes must satisfy two contradictory constraints: to be representative (in the sense of the metric) and of cardinality as small as possible. The problem thus reduces to that of inferring an epsilon-network of minimal cardinality. We treat it in all its generality by discussing the spaces on which the data are defined. Our method is based on hierarchical agglomerative classification with as linkage the radius of the minimum balls enclosing the points of each subset. Applied to our libraries, this approach reduced their size by a factor of 4, and our docking computation time by the same amount, while improving their reliability.Finally, to overcome the problem posed by the pairwise superimposition of structures, we used a representation of the fragments in internal coordinates, allowing to reduce further the computation time for the creation of libraries
Bérenger, François. "Nouveaux logiciels pour la biologie structurale computationnelle et la chémoinformatique." Thesis, Paris, CNAM, 2016. http://www.theses.fr/2016CNAM1047/document.
Full textThis thesis introduces five software useful in three different areas : parallel and distributed computing, computational structural biology and chemoinformatics. The software from the parallel and distributed area is PAR. PAR allows to execute independent experiments in a parallel and distributed way. The software for computational structural biology are Durandal, EleKit and Fragger. Durandal exploits the propagation of geometric constraints to accelerate the exact clustering algorithm for protein models. EleKit allows to measure the electrostatic similarity between a chemical molecule and the protein it is designed to replace at a protein-protein interface. Fragger is a fragment picker able to select protein fragments in the whole protein data-bank. Finally, the chemoinformatics software is ACPC. ACPC encodes in a rotation-translation invariant way a chemical molecule in any or a combination of three chemical spaces (electrostatic, steric or hydrophobic). ACPC is a ligand-based virtual screening tool supporting consensus queries, query molecule annotation and multi-core computers
Lombard, Valentin. "Geometric deep manifold learning combined with natural language processing for protein movies." Electronic Thesis or Diss., Sorbonne université, 2024. http://www.theses.fr/2024SORUS379.
Full textProteins play a central role in biological processes, and understanding how they deform and move is essential to elucidating their functional mechanisms. Despite recent advances in high-throughput technologies, which have broadened our knowledge of protein structures, accurate prediction of their various conformational states and motions remains a major challenge. We present two complementary approaches to address the challenge of understanding and predicting the full range of protein conformational variability. The first approach, Dimensionality Analysis for protein Conformational Exploration (DANCE) for a systematic and comprehensive description of protein families conformational variability. DANCE accommodates both experimental and predicted structures. It is suitable for analyzing anything from single proteins to superfamilies. Employing it, we clustered all experimentally resolved protein structures available in the Protein Data Bank into conformational collections and characterized them as sets of linear motions. The resource facilitates access and exploitation of the multiple states adopted by a protein and its homologs. Beyond descriptive analysis, we assessed classical dimensionality reduction techniques for sampling unseen states on a representative benchmark. This work improves our understanding of how proteins deform to perform their functions and opens ways to a standardized evaluation of methods designed to sample and generate protein conformations. The second approach relies on deep learning to predict continuous representations of protein motion directly from sequences, without the need for structural data. This model, SeaMoon, uses protein language model (pLM) embeddings as inputs to a lightweight convolutional neural network with around 1 million trainable parameters. SeaMoon achieves a success rate of 40% when evaluated against around 1,000 collections of experimental conformations, capturing movements beyond the reach of traditional methods such as normal mode analysis, which relies solely on 3D geometry. In addition, SeaMoon generalizes to proteins that have no detectable sequence similarity with its training set and can be easily retrained with updated pLMs. These two approaches offer a unified framework for advancing our understanding of protein dynamics. DANCE provides a detailed exploration of protein movements based on structural data, while SeaMoon demonstrates the potential of sequence-based deep learning models to capture complex movements without relying on explicit structural information. Together, they pave the way for a more comprehensive understanding of protein conformational variability and its role in biological function