Dissertations / Theses on the topic 'In silico drug prediction'
To see the other types of publications on this topic, follow the link: In silico drug prediction.
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 'In silico drug prediction.'
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.
1
Ahlin, Gustav. "In vitro and in silico prediction of drug-drug interactions with transport proteins." Doctoral thesis, Uppsala : Acta Universitatis Upsaliensis, 2009. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-107492.
Full text
2
Carrió, Gaspar Pau 1982. "Development of advanced strategies for the prediction of toxicity endpoints in drug development." Doctoral thesis, Universitat Pompeu Fabra, 2015. http://hdl.handle.net/10803/328418.
Full textAbstract:
Safety concerns are one of the main causes of drug attrition. In these events, the moment at which the drug toxic effects are discovered changes dramatically the importance of the finding; discarding a valuable candidate at clinical testing stages means wasting years of efforts and huge economicinvestments. Even more dramatic is the discovery of toxic effect at post marketing stages, when the drug could have already produced severe side effects on a number of patients.
For these reason there is a pushing need of developing methods able to assess the safety of drug candidates at early stages of development. Among these, in silico methods have many advantages, like not even requiring the availability of the compound, not wasting any quantity of it in case it has been already synthesized, being fast, cheap and make no use of animal testing. Unfortunately, in silico prediction methods of toxicity endpoints do not perform always as expected. The reasons are still under debate, but likely reasons are the complexity of the biological phenomena under study and the large structural diversity of the drug candidates, among others.
The aim of this thesis is to improve currently used in silico prediction methods for their application to biological endpoints of interest in drug development, with a special emphasis to toxicological endpoints. Here, we report a novel general methodology called ADAN (Applicability Domain Analysis) for assessing the reliability of drug property predictions obtained by in silico methods. Furthermore, we proposed a unifying strategy for the use of in silico predictive methods in this field, defining rational criteria for the application of a whole spectrum of methods; from structural alerts to global QSAR models, including read across and local models. The usefulness of all the proposed methodologies is tested using a systematic analysis on representative datasets, obtaining good results that confirm their validity.La manca de seguretat és una de les raons principals per la qual els candidats a fàrmacs són descartats. La fase en què els possibles efectes tòxics són identificats és crítica: descartar un candidat en fase clínica implica la pèrdua d'anys d'esforços i enormes inversions econòmiques. Encara pitjor és identificar efectes tòxics un cop el fàrmac està comercialitzat, quan es poden haver produït greus efectes secundaris en pacients. Per aquestes raons hi ha la necessitat de desenvolupar mètodes capaços d'avaluar la seguretat dels candidats a fàrmacs en les primeres etapes. Entre aquests, els mètodes in silico tenen molts avantatges, com no requerir la disponibilitat del compost, no perdre cap quantitat en cas que ja s'hagi sintetitzat, ser ràpid, econòmic i no fer ús de l'experimentació amb animals. Per desgràcia, els mètodes de predicció in silico aplicats a criteris d'avaluació de toxicitat no produeixen els resultats adequats. Les raons són objecte de debat, però raons probables són la complexitat dels fenòmens biològics en estudi i la gran diversitat estructural els fàrmacs candidats, entre d'altres. L'objectiu d'aquesta tesi és millorar els mètodes de predicció in silico emprats en l’avaluació de criteris d'interès en el desenvolupament de fàrmacs amb especial èmfasi en els de toxicitat. Presentem una nova metodologia general anomenada ADAN (Applicability Domain Analysis) per avaluar la fiabilitat de les prediccions obtingudes amb mètodes in silico. A més, proposem una estratègia unificada de l’ús de mètodes de predicció in silico emprats en aquest camp; com alertes estructurals, read-across, QSAR local i global. La estratègia incorpora criteris racionals per la seva utilització. Els bons resultats obtinguts amb dades representatives confirmen la validesa de les metodologies.
3
Vildhede, Anna. "In vitro and in silico Predictions of Hepatic Transporter-Mediated Drug Clearance and Drug-Drug Interactions in vivo." Doctoral thesis, Uppsala universitet, Institutionen för farmaci, 2015. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-241376.
Full textAbstract:
The liver is the major detoxifying organ, clearing the blood from drugs and other xenobiotics. The extent of hepatic clearance (CL) determines drug exposure and hence, the efficacy and toxicity associated with exposure. Drug-drug interactions (DDIs) that alter the hepatic CL may cause more or less severe outcomes, such as adverse drug reactions. Accurate predictions of drug CL and DDI risk from in vitro data are therefore crucial in drug development. Liver CL depends on several factors including the activities of transporters involved in the hepatic uptake and efflux. The work in this thesis aimed at developing new in vitro and in silico methods to predict hepatic transporter-mediated CL and DDIs in vivo. Particular emphasis was placed on interactions involving the hepatic uptake transporters OATP1B1, OATP1B3, and OATP2B1. These transporters regulate the plasma concentration-time profiles of many drugs including statins. Inhibition of OATP-mediated transport by 225 structurally diverse drugs was investigated in vitro. Several novel inhibitors were identified. The data was used to develop in silico models that could predict OATP inhibitors from molecular structure. Models were developed for static and dynamic predictions of in vivo transporter-mediated drug CL and DDIs. These models rely on a combination of in vitro studies of transport function and mass spectrometry-based quantification of protein expression in the in vitro models and liver tissue. By providing estimations of transporter contributions to the overall hepatic uptake/efflux, the method is expected to improve predictions of transporter-mediated DDIs. Furthermore, proteins of importance for hepatic CL were quantified in liver tissue and isolated hepatocytes. The isolation of hepatocytes from liver tissue was found to be associated with oxidative stress and degradation of transporters and other proteins expressed in the plasma membrane. This has implications for the use of primary hepatocytes as an in vitro model of the liver. Nevertheless, by taking the altered transporter abundance into account using the method developed herein, transport function in hepatocyte experiments can be scaled to the in vivo situation. The concept of protein expression-dependent in vitro-in vivo extrapolations was illustrated using atorvastatin and pitavastatin as model drugs.
4
Bergström, Christel A. S. "Computational and Experimental Models for the Prediction of Intestinal Drug Solubility and Absorption." Doctoral thesis, Uppsala University, Department of Pharmacy, 2003. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-3593.
Full textAbstract:
New effective experimental techniques in medicinal chemistry and pharmacology have resulted in a vast increase in the number of pharmacologically interesting compounds. However, the number of new drugs undergoing clinical trial has not augmented at the same pace, which in part has been attributed to poor absorption of the compounds.
The main objective of this thesis was to investigate whether computer-based models devised from calculated molecular descriptors can be used to predict aqueous drug solubility, an important property influencing the absorption process. For this purpose, both experimental and computational studies were performed. A new small-scale shake flask method for experimental solubility determination of crystalline compounds was devised. This method was used to experimentally determine solubility values used for the computational model development and to investigate the pH-dependent solubility of drugs. In the computer-based studies, rapidly calculated molecular descriptors were used to predict aqueous solubility and the melting point, a solid state characteristic of importance for the solubility. To predict the absorption process, drug permeability across the intestinal epithelium was also modeled.
The results show that high quality solubility data of crystalline compounds can be obtained by the small-scale shake flask method in a microtiter plate format. The experimentally determined pH-dependent solubility profiles deviated largely from the profiles predicted by a traditionally used relationship, highlighting the risk of data extrapolation. The in silico solubility models identified the non-polar surface area and partitioned total surface areas as potential new molecular descriptors for solubility. General solubility models of high accuracy were obtained when combining the surface area descriptors with descriptors for electron distribution, connectivity, flexibility and polarity. The used descriptors proved to be related to the solvation of the molecule rather than to solid state properties. The surface area descriptors were also valid for permeability predictions, and the use of the solubility and permeability models in concert resulted in an excellent theoretical absorption classification. To summarize, the experimental and computational models devised in this thesis are improved absorption screening tools applicable to the lead optimization in the drug discovery process.
5
Carlert, Sara. "Investigation and Prediction of Small Intestinal Precipitation of Poorly Soluble Drugs : a Study Involving in silico, in vitro and in vivo Assessment." Doctoral thesis, Uppsala universitet, Institutionen för farmaci, 2012. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-178053.
Full textAbstract:
The main objectives of the present project were to increase the understanding of small intestinal precipitation of poorly soluble pharmaceutical drugs, investigate occurrence of crystalline small intestinal precipitation and effects of precipitation on absorption. The aim was to create and evaluate methods of predicting crystalline small intestinal drug precipitation using in vivo, in vitro and in silico models. In vivo small intestinal precipitation from highly supersaturated solutions of two weakly basic model drugs, AZD0865 and mebendazole, was investigated in humans and canine models. Potential precipitation of AZD0865 was investigated by examining dose dependent increases in human maximum plasma concentration and total exposure, which turned out to be dose linear over the range investigated, indicating no significant in vivo precipitation. The small intestinal precipitation of mebendazole was investigated from drug concentrations and amount of solid drug present in dog jejunum as well as through the bioavailability after direct duodenal administration in dogs. It was concluded that mebendazole small intestinal precipitation was limited, and that intestinal supersaturation was measurable for up to 90 minutes. In vitro precipitation methods utilizing simulated or real fasted gastric and intestinal fluids were developed in order to simulate the in vivo precipitation rate. The methods overpredicted in vivo precipitation when absorption of drug was not simulated. An in vitro-in silico approach was therefore developed, where the in vitro method was used for determining the interfacial tension (γ), necessary for describing crystallization in Classical Nucleation Theory (CNT). CNT was evaluated using a third model drug, bicalutamide, and could successfully describe different parts of the crystallization process of the drug. CNT was then integrated into an in silico absorption model. The in vivo precipitation results of AZD0865 and mebendazole were well predicted by the model, but only by allowing the fundamental constant γ to vary with concentration. Thus, the in vitro-in silico approach could be used for small intestinal precipitation prediction if the in vitro concentration closely matched in vivo small intestinal concentrations.
6
Weston, Anne. "In silico prediction of protein function." Thesis, King's College London (University of London), 2004. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.412922.
Full text
7
Sandelin, Albin. "In silico prediction of CIS-regulatory elements /." Stockholm, 2004. http://diss.kib.ki.se/2004/91-7349-879-3/.
Full text
8
Cereto, Massagué Adrià. "Development of tools for in silico drug discovery." Doctoral thesis, Universitat Rovira i Virgili, 2017. http://hdl.handle.net/10803/454678.
Full textAbstract:
El cribratge virtual és un mètode quimioinformàtic que consisteix en cribrar molècules bioactives de grans bases de dades de molècules petites. Això permet als investigadors d’estalviar-se el cost de provar experimentalment cents o milers de compostos candidats, reduïnt-ne el nombre fins a quantitats manejables. Per a la validació dels mètodes de cribratge virtual calen biblioteques de molècules cimbell. El programari DecoyFinder fou desenvolupat com a aplicació gràfica de fàcil ús per a la construcció de biblioteques de molècules cimbell, i fou posteriorment ampliat amb les troballes de recerca posterior sobre la construcció i rendiment de biblioteques de molècules cimbell.
El Protein Data Bank (PDB) és molt útil perquè proporciona estructures tridimensionals per a complexos proteïna-lligand, i per tant, informació sobre com interactuen. Pels mètodes de cribratge virtual que en depenen, n’és extremadament important la seva fiabilitat. El VHELIBS fou desenvolupat com a eina per a inspeccionar i identificar, fàcilment i intuitiva, les estructures fiables del PDB, basant-se en com de bo n’és l’encaix amb els seus corresponents mapes de densitat electrònica.
Mentre que el cribratge virtual prova de trobar noves molècules bioactives per determinades dianes, l’enfoc invers també s’empra: arran d’una molècula, cercar-ne dianes amb activitat biològica no documentada. Aquest cribratge invers és conegut en anglès com a “in silico target fishing”, o pesca de dianes “in silico”, i és especialment útil a l’àmbit de la reutilització de fàrmacs En començar aquesta tesi, no hi havia cap plataforma de “target fishing” de lliure accés, i tot i que durant els anys se n’han desenvolupat algunes, en tots els casos la seva predicció de bioactivitat és qualitativa. Per això es desenvolupà una plataforma pròpia de “target fishing” de lliure accés, amb la implementació d’un nou mètode que proporciona la primera predicció quantitativa de bioactivitat per aquest tipus de plataforma.El cribado virtual es un método quimioinformático que consiste en la criba de moléculas bioactivas de grandes bases de datos de moléculas pequeñas. Esto permite a los investigadores ahorrarse el coste de probar experimentalmente cientos o miles de compuestos candidatos, reduciéndolos hasta cantidades manejables. Para la validación de los métodos de cribado virtual hacen falta bibliotecas de moléculas señuelo. El software DecoyFinder fue desarrollado como aplicación gráfica de fácil uso para la construcción de bibliotecas de moléculas señuelo, y fue posteriormente ampliado con los hallazgos de investigación posterior sobre la construcción i rendimiento de bibliotecas de moléculas señuelo. El Protein Data Bank (PDB) es muy útil porque proporciona estructuras tridimensionales para complejos proteina-ligando, y por tanto, información sobre como interactúan. Para los métodos de cribado virtual que dependen de ellas, es extremadamente importante su fiabilidad. VHELIBS fue desarrollado como herramienta para inspeccionar e identificar, fácil e intuitivamente, las estructuras fiables del PDB, basándose en como de bueno es su encaje con sus correspondientes mapas de densidad electrónica. Mientras que el cribado virtual intenta encontrar nuevas moléculas bioactivas para determinadas dianas, el enfoque inverso también se utiliza: a partir de una molécula, buscar dianas donde presente actividad biológica no documentada. Este cribado inverso es conocido en inglés como “in silico target fishing”, o pesca de dianas “in silico”, y es especialmente útil en el ámbito de la reutilización de fármacos. Al comenzar esta tesis, no había ninguna plataforma de “target fishing” de libre acceso, y aunque durante los años se han desarrollado algunas, en todos los casos su predicción de bioactividad es cualitativa. Por eso se desarrolló una plataforma propia de “target fishing” de libre acceso, con la implementación de un nuevo método que proporciona la primera predicción cuantitativa de bioactividad para este tipo de plataforma.
Virtual screening is a cheminformatics method that consists of screening large small-molecule databases for bioactive molecules. This enables the researcher to avoid the cost of experimentally testing hundreds or thousands of compounds by reducing the number of candidate molecules to be tested to manageable numbers. For their validation, virtual screening approaches need decoy molecule libraries. DecoyFinder was developed as an easy to use graphical application for decoy library building, and later updated after some research into decoy library building and their performance when used for 2D similarity approaches. The Protein Data Bank (PDB) is very useful because it provides 3D structures for protein-ligand complexes and, therefore, information on how certain ligands bind and interact with their targets. For virtual screening apporaches relying on these structures, it is of the utmost importance that the data available on the PDB for the ligand and its binding site are reliable. VHELIBS was developed as a tool to easily and intuitively inspect and identify reliable PDB structures based on the goodness of fitting between ligands and binding sites and their corresponding electron density map. While virtual screening aims to find new bioactive molecules for certain targets, the opposite approach is also used: starting from a given molecule, to search for a biological target for which it presents previously undocumented bioactivity. This reverse screening is known as in silico or computational target fishing or reverse pharmacognosy, and it is specially useful for drug repurposing or repositioning. When this thesis was started, there were no freely available target fishing platforms, but some have been developed during the years. However, they are qualitative in the nature of their activity prediction, and thus we set out to develop a freely accessible target fishing web service implementing a novel method which provides the first quantitative activity prediction: Anglerfish.
9
Acoca, Stephane. "In silico methods in drug discovery and development." Thesis, McGill University, 2012. http://digitool.Library.McGill.CA:80/R/?func=dbin-jump-full&object_id=110376.
Full textAbstract:
Computational drug design methods have become increasingly invaluable in the drug discovery and development process. Throughout this thesis will be described the development and application of methods that are used at every stage of the drug discovery and development pipeline. In Chapter 2 will take a look at the use computational methods towards the understanding and development of two novel Bcl-2 inhibitors, Obatoclax and ABT-737, being developed for the treatment of Cancer. The study proposes certain mechanisms through which ABT-737 displays selectivity towards certain targets within the Bcl-2 family. Additionally, we propose a binding mode for Obatoclax which is in accordance with experimental data. The following Chapter addresses the use of virtual screening for the identification of novel lead compounds. Trypanosoma brucei RNA Editing Ligase 1 was chosen as the target for the development of treatments against Trypanosoma infections and C35, a potent novel inhibitor of the enzyme, was identified. Furthermore, our research shows that the action of C35 extends to inhibition of several critical enzyme activities required for the RNA editing process as well as compromising the integrity of the multiprotein complex which carries it out. The following Chapter takes a look at the use of mass spectrometry data in order to expedite discovery of bioactive compounds in natural products. We developed an algorithm which analyses MS/MS data in order to derive the Molecular Formula of the compound. The novel algorithm obtained a 95% success rate on a test set of 91 compounds. The last Chapter of the thesis explores the use of molecular dynamics to generate a conformational ensemble of targets for virtual screening. Conformational ensembles were generated for a target test set taken from the Directory for Useful Decoys. The results showed that molecular dynamics-based conformational ensembles provided remarkable improvements on 2 of the targets tested due to the enhanced capacity to properly dock compounds in otherwise restricted structures. The last Chapter of the thesis is a general discussion on the work of the thesis and a proposal on how all can be integrated within the drug discovery and development pipeline.Les méthodes the modélisation sont devenues un outil inestimable dans le processus de découverte et de développement de nouveaux médicaments. Au cours de cette thèse va être décrit le développement et l'application de méthodes utilisés à chaque stage de la découverte et du développement de produits pharmaceutiques. Le Chapitre 2 est un aperçu sur l'utilisation de méthodes computationnelles vers le développement de deux nouveaux inhibiteurs des protéines Bcl-2, Obatoclax et ABT-737, en développement pour le traitement du Cancer. L'étude propose certains mécanismes d'ABT-737 qui expliquent ca sélectivité envers les membres de la famille Bcl-2. De plus, nous proposons un mécanisme d'attachement pour Obatoclax qui conforme aux données expérimentales. Le Chapitre suivant adresse l'utilisation du dépistage virtuel pour l'identification de nouvelles molécules mère. La Ligase de l'Edition d'ARN du Trypanosoma brucei a été choisie comme cible pour le développement de traitements contre des infections dû au Trypanosome et C35 a été identifié comme nouvel inhibiteur de l'enzyme. En outre, notre recherche démontre que l'action de C35 s'étends a l'inhibition de plusieurs enzymes nécessaires pour le mécanisme d'édition de l'ARN en plus de compromettre l'intégrité du complexe multi-protéinique qui l'effectue. Le Chapitre suivant prends regard a l'utilisation de donnes dérivant de la spectrométrie de masse pour but d'accélérer la découverte de molécules bioactives venant de sources naturelles. Nous avons développé un algorithme qui analyse les données MS/MS pour but de dériver la formule moléculaire du composé. Le nouvel algorithme a obtenu un taux de succès s'élevant à 95% sur un ensemble test de 91 molécules. Le dernier Chapitre de la thèse explore l'utilisation de simulations de dynamique moléculaire pour générer en ensemble conformationel de protéines cible pour son utilisation dans le dépistage virtuel. Les ensembles conformationel ont étés généré pour une série test obtenu d'un répertoire attitré 'Directory for Useful Decoys'. Les résultats démontrent que les ensembles conformationel dérivés de la dynamique moléculaire ont apporté des améliorations remarquables sur deux des cibles testées dû à une capacité accrue de placement approprié des molécules dans un site qui est autrement très restreint. Le dernier Chapitre de cette thèse est une discussion générale sur le travail accomplie et une proposition sur la manière dont tous les éléments sont intégrer dans un protocole de découverte et de développement de produits pharmaceutiques.
10
Kundu, Kousik [Verfasser], and Rolf [Akademischer Betreuer] Backofen. "In Silico Prediction of Modular Domain-Peptide Interactions." Freiburg : Universität, 2015. http://d-nb.info/1115861883/34.
Full text
11
Dörr, Alexander [Verfasser]. "In Silico Approaches for Polypharmacological Drug Design / Alexander Dörr." München : Verlag Dr. Hut, 2018. http://d-nb.info/1156510287/34.
Full text
12
Dickson, Callum. "In silico modelling of membranes and drug membrane interactions." Thesis, Imperial College London, 2014. http://hdl.handle.net/10044/1/25070.
Full textAbstract:
A new all-atom force field for the simulation of phospholipid bilayers using the AMBER molecular dynamics package has been developed, which is compatible with other AMBER protein, nucleic acid, carbohydrate and small molecule force fields. The force field has been validated by simulating bilayers of six different lipid types, finding favourable comparison to experiment for properties such as area per lipid, volume per lipid, bilayer thickness, NMR order parameters, scattering data, and lipid lateral diffusion. The modular nature of this force field allows numerous combinations of head and tail groups to create different lipid types, enabling the easy insertion of new lipid species. The lipid bilayer model has then been applied to the study of the interaction between radioimaging agents and membranes in an effort to understand the phenomena of non-specific binding, which remains poorly understood yet of serious detrimental consequence to the development of new imaging tracers. The effect of different concentrations of imaging agent on a homogeneous membrane has been examined using unbiased simulations, whilst the permeability coefficient of each imaging agent through a membrane has been calculated using biased simulations. It is found that radiotracers with low non-specific binding must adopt a certain orientation to cross the head group region of a membrane - this requirement may act as a barrier to membrane entry. Furthermore, once partitioned into the membrane, simulations predict that those radiotracers displaying a high degree of non-specific binding act to order lipid tail groups to a greater extent than those with low non-specific binding, reducing the permeability of the membrane and possibly acting to 'trap' radiotracer in the membrane. These simulations also predict that non-specific binding is not related to radiotracer membrane permeability through a homogeneous bilayer.
13
Vernikos, Georgios. "In silico prediction of genomic islands in microbial genomes." Thesis, University of Cambridge, 2008. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.612465.
Full text
14
Thovarai, Vishal. "In silico drug design of potential novel anti malarial agents /." Online version of thesis, 2009. http://hdl.handle.net/1850/8689.
Full text
15
Elisée, Eddy. "Towards in silico prediction of mutations related to antibiotic resistance." Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLS350.
Full textAbstract:
La résistance aux antibiotiques est une menace sérieuse pour la santé publique. En effet, si on ne change pas rapidement notre consommation excessive d'antibiotiques, la situation actuelle va se dégrader jusqu'à basculer dans une ère dite "post-antibiotique", dans laquelle plus aucun antibiotique ne sera efficace contre les infections microbiennes. Bien que ce phénomène de résistance apparaît naturellement, l'utilisation abusive d'antibiotiques accélère le processus. De plus, la présence de pathogènes multi-résistants neutralise l'effet des traitements existants et dans le cas de chirurgies courantes (césariennes, transplantations d'organe...), la situation peut rapidement s'aggraver voire devenir mortelle. C'est pourquoi des directives, émanant des autorités sanitaires, doivent être mises en place afin de contrôler l'utilisation des médicaments, et ce, à tous les niveaux de la société, des individus au secteur agricole en passant par les professionnels de santé et les industries pharmaceutiques. Le monde de la recherche scientifique, quant à elle, doit trouver des nouvelles stratégies pour enrayer la propagation de la résistance. Dans ce contexte, cette thèse a pour objectif le développement d'une méthode de prédiction, par calculs d'énergie libre, des mutations de β-lactamases favorables à l'hydrolyse des β-lactames. Ces travaux méthodologiques ont donc conduit au développement : (1) de nouveaux paramètres pour les enzymes à zinc, implémentés dans le champ de force OPLS-AA et validés par des simulations de dynamique moléculaire sur un panel de métalloenzymes représentatives, (2) d'un protocole de paramétrisation de ligands covalents pour étudier le comportement de certains β-lactames dans CMY-136, une nouvelle β-lactamase caractérisée au laboratoire, et (3) d'un protocole de calcul d'énergie libre évalué au moyen de compétitions internationales de prédiction. Ce dernier a ensuite été utilisé pour tenter d'expliquer pourquoi la carbamylation de la sérine catalytique n'a pas lieu dans certaines oxacillinases. Au travers de ces travaux, nous avons pu améliorer significativement notre approche computationnelle et désormais tout est en place pour une exploration exhaustive des mutations possibles dans les β-lactamasesAntibiotic resistance is a global concern threatening worldwide health. Indeed, if we don't change our overconsumption of antibiotics, the current situation could worsen until a "post-antibiotic" era in which existing treatment would be ineffective against microbial infections. Despite the natural occurrence of antibiotic resistance, the misuse of antibiotics is speeding up the process. Furthermore, presence of multi-resistant pathogens negates the effect of modern treatments and usual surgeries (caesarean sections, organ transplantations...) might be riskier in the future, or even lethal. That's why, common guidelines have to be edicted by health authorities in order to control antibiotic use at every level of society, from individuals to healthcare industry including health professionals and agriculture sector. As for scientific research, new strategies have to be considered in order to limit the spread of antibiotic resistance. In that context, the presented thesis aimed at developing a protocol to predict, by free energy calculations, β-lactamase mutations which could promote the hydolysis of β-lactams antibiotics. In order to achieve that, we developed several methodological approaches including: (1) new parameters for zinc enzymes implemented in OPLS-AA force field and thereafter validated using molecular dynamics simulations of representative zinc-containing metalloenzymes, (2) a protocol to parameterize covalent ligands in order to analyze the dynamical behavior of some β-lactams in CMY-136, a novel β-lactamase recently characterized in our laboratory, and (3) a pmx-based free energy protocol. The latter was also assessed through several international blinded prediction challenges, and finally used to find out why carbamylation of the catalytic serine is not observed in certain OXA enzymes. Throughout this work, we made significant improvements in our protocol, and now everything is in place for an exhaustive prediction of possible mutations in β-lactamases
16
Hamad, Mustafa. "In-Silico Prediction of the Physical Performance of Pharmaceutical Crystals." Thesis, Curtin University, 2021. http://hdl.handle.net/20.500.11937/86652.
Full textAbstract:
Knowledge of the hardness of pharmaceutical crystals expedites the large-scale manufacturing of drug tablets. This research involved the development and testing of computational methods to simulate the various deformations of crystals that lead to the calculation of hardness. Both methods were applied to a range of materials and provided a consistent ranking of the slip systems with experiment, a detailed atomistic description of the deformation mechanisms, and the calculation of ideal shear strength.
17
Elkaïm, Judith. "Drug design in silico : criblage virtuel de protéines à visée thérapeutique." Thesis, Bordeaux 1, 2011. http://www.theses.fr/2011BOR14444/document.
Full textAbstract:
Les processus qui mènent à la découverte de nouveaux médicaments sont longs et fastidieux, et les taux de succès sont relativement faibles. L’identification de candidats par le biais de tests expérimentaux s’avère coûteuse, et nécessite de connaître en profondeur les mécanismes d'action de la protéine visée afin de mettre en place des essais efficaces. Le criblage virtuel peut considérablement accélérer ces processus en permettant une évaluation rapide de chimiothèques de plusieurs milliers de molécules afin de déterminer lesquelles sont les plus susceptibles de se lier à une cible. Ces dernières années ont ainsi été témoins de quelques success stories dans ce domaine.Le premier objectif de ce travail était de comparer différents outils et stratégies couramment utilisés dans le criblage virtuel “structure-based”, puis de les appliquer à des cibles protéiques à visée thérapeutique, en particulier dans le cadre du cancer.La protéine kinase GSK3 et un test set de ligands connus ont servi de modèle pour différentes études méthodologiques ayant pour but d’évaluer les programmes de docking et de scoring à notre disposition. En particulier, l’utilisation de plusieurs structures relaxées du récepteur ou l’insertion de torsions sur certains résidus du site actif pendant le docking ont permis d’évaluer l’influence de la flexibilité de la protéine. L’utilité et la pertinence d’outils permettant de générer automatiquement les structures 3D des ligands et de méthodes de consensus scoring ont également été étudiées.Un criblage virtuel de la Pontine, une ATPase impliquée dans la croissance tumorale pour laquelle aucun inhibiteur n’était connu, a permis la sélection de candidats issus de banques de données commerciales. Ces molécules ont été testées dans un essai enzymatique par le biais d’une collaboration, et quatre d’entre elles se sont révélées capable d’inhiber l’activité ATPase de la Pontine. Le criblage de bases de ligands synthétisés et imaginés dans l’équipe a également fourni un inhibiteur original. Au contraire, l’étude de la sPLA2-X humaine, une phospholipase dont l’activité catalytique est dépendante d’un atome de Ca2+ localisé au sein du site actif, a montré les limites de nos outils de docking qui n’ont pas été capables de gérer cet ion métallique et mis en évidence la nécessité de mettre en place d’autres outilsThe process of drug discovery is long and tedious. Besides, it is relatively inefficient in terms of hit rate. The identification of candidates through experimental testing is expensive and requires extensive data on the mechanisms of the target protein in order to develop efficient assays. Virtual screening can considerably accelerate the process by quickly evaluating large databases of compounds and determining the most likely to bind to a target. Some success stories have emerged in the field over the last few years.The objectives of this work were first, to compare common tools and strategies for structure-based virtual screening, and second, to apply those tools to actual target proteins implied notably in carcinogenesis.In order to evaluate the docking and scoring programs available, the protein kinase GSK3 and a test set of known ligands were used as a model to perform methodological studies. In particular the influence of the flexibility of the protein was explored via relaxed structures of the receptor or the insertion of torsions on the side chains of residues located in the binding site. Studies concerning the automatic generation of 3D structures for the ligands and the use of consensus scoring also provided insights on the usability of these tools while performing a virtual screening.Virtual screening of the human protein Pontin, an ATPase implied in tumor cell growth for which no inhibitors were known, allowed the prioritization of compounds from commercial databases. These compounds were tested in an enzymatic assay via a collaboration, and led to the identification of four molecules capable of inhibiting the ATPase activity of Pontin. Additional screens of in-house oriented databases also provided at least one innovative inhibitor for this protein. On the contrary, a study of the human PLA2-X, a phospholipase that requires a Ca2+ atom to bind to its active site in order to catalyze the hydrolysis of its substrate, revealed the limits of our docking tools that could not handle the metal ion and the need for new tools
18
Boateng, Rita Afriyie. "In silico characterization of plasmodial transketolases as potential malaria drug target." Thesis, Rhodes University, 2018. http://hdl.handle.net/10962/63540.
Full text
19
Yella, Jaswanth. "Machine Learning-based Prediction and Characterization of Drug-drug Interactions." University of Cincinnati / OhioLINK, 2018. http://rave.ohiolink.edu/etdc/view?acc_num=ucin154399419112613.
Full text
20
Kastenmüller, Gabriele. "In silico prediction and comparison of metabolic capabilities in sequenced genomes /." München : Hut, 2009. http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&doc_number=018929163&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA.
Full text
21
Lin, Frank Po-Yen Centre for Health Informatics Faculty of Medicine UNSW. "In silico virulence prediction and virulence gene discovery of Streptococcus agalactiae." Awarded By:University of New South Wales. Centre for Health Informatics, 2009. http://handle.unsw.edu.au/1959.4/44382.
Full textAbstract:
Physicians frequently face challenges in predicting which bacterial subpopulations are likely to cause severe infections. A more accurate prediction of virulence would improve diagnostics and limit the extent of antibiotic resistance. Nowadays, bacterial pathogens can be typed with high accuracy with advanced genotyping technologies. However, effective translation of bacterial genotyping data into assessments of clinical risk remains largely unexplored. The discovery of unknown virulence genes is another key determinant of successful prediction of infectious disease outcomes. The trial-and-error method for virulence gene discovery is time-consuming and resource-intensive. Selecting candidate genes with higher precision can thus reduce the number of futile trials. Several in silico candidate gene prioritisation (CGP) methods have been proposed to aid the search for genes responsible for inherited diseases in human. It remains uninvestigated as to how the CGP concept can assist with virulence gene discovery in bacterial pathogens. The main contribution of this thesis is to demonstrate the value of translational bioinformatics methods to address challenges in virulence prediction and virulence gene discovery. This thesis studied an important perinatal bacterial pathogen, group B streptococcus (GBS), the leading cause of neonatal sepsis and meningitis in developed countries. While several antibiotic prophylactic programs have successfully reduced the number of early-onset neonatal diseases (infections that occur within 7 days of life), the prevalence of late-onset infections (infections that occur between 7??30 days of life) remained constant. In addition, the widespread use of intrapartum prophylactic antibiotics may introduce undue risk of penicillin allergy and may trigger the development of antibiotic-resistant microorganisms. To minimising such potential harm, a more targeted approach of antibiotic use is required. Distinguish virulent GBS strains from colonising counterparts thus lays the cornerstone of achieving the goal of tailored therapy. There are three aims of this thesis: 1. Prediction of virulence by analysis of bacterial genotype data: To identify markers that may be associated with GBS virulence, statistical analysis was performed on GBS genotype data consisting of 780 invasive and 132 colonising S. agalactiae isolates. From a panel of 18 molecular markers studied, only alp3 gene (which encodes a surface protein antigen commonly associated with serotype V) showed an increased association with invasive diseases (OR=2.93, p=0.0003, Fisher??s exact test). Molecular serotype II (OR=10.0, p=0.0007) was found to have a significant association with early-onset neonatal disease when compared with late-onset diseases. To investigate whether clinical outcomes can be predicted by the panel of genotype markers, logistic regression and machine learning algorithms were applied to distinguish invasive isolates from colonising isolates. Nevertheless, the predictive analysis only yielded weak predictive power (area under ROC curve, AUC: 0.56??0.71, stratified 10-fold cross-validation). It was concluded that a definitive predictive relationship between the molecular markers and clinical outcomes may be lacking, and more discriminative markers of GBS virulence are needed to be investigated. 2. Development of two computational CGP methods to assist with functional discovery of prokaryotic genes: Two in silico CGP methods were developed based on comparative genomics: statistical CGP exploits the differences in gene frequency against phenotypic groups, while inductive CGP applies supervised machine learning to identify genes with similar occurrence patterns across a range of bacterial genomes. Three rediscovery experiments were carried out to evaluate the CGP methods: a) Rediscovery of peptidoglycan genes was attempted with 417 published bacterial genome sequences. Both CGP methods achieved their best AUC >0.911 in Escherichia coli K-12 and >0.978 Streptococcus agalactiae 2603 (SA-2603) genomes, with an average improvement in precision of >3.2-fold and a maximum of >27-fold using statistical CGP. A median AUC of >0.95 could still be achieved with as few as 10 genome examples in each group in the rediscovery of the peptidoglycan metabolism genes. b) A maximum of 109-fold improvement in precision was achieved in the rediscovery of anaerobic fermentation genes. c) In the rediscovery experiment with genes of 31 metabolic pathways in SA-2603, 14 pathways achieved an AUC >0.9 and 28 pathways achieved AUC >0.8 with the best inductive CGP algorithms. The results from the rediscovery experiments demonstrated that the two CGP methods can assist with the study of functionally uncategorised genomic regions and the discovery of bacterial gene-function relationships. 3. Application of the CGP methods to discover GBS virulence genes: Both statistical and inductive CGP were applied to assist with the discovery of unknown GBS virulence factors. Among a list of hypothetical protein genes, several highly-ranked genes were plausibly involved in molecular mechanisms in GBS pathogenesis, including several genes encoding family 8 glycosyltransferase, family 1 and family 2 glycosyltransferase, multiple adhesins, streptococcal neuraminidase, staphylokinase, and other factors that may have roles in contributing to GBS virulence. Such genes may be candidates for further biological validation. In addition, the co-occurrence of these genes with currently known virulence factors suggested that the virulence mechanisms of GBS in causing perinatal diseases are multifactorial. The procedure demonstrated in this prioritisation task should assist with the discovery of virulence genes in other pathogenic bacteria.
22
Guest, Eleanor. "Assessment of algorithms for the prediction of metabolic drug-drug interactions." Thesis, University of Manchester, 2011. https://www.research.manchester.ac.uk/portal/en/theses/assessment-of-algorithms-for-the-prediction-of-metabolic-drugdrug-interactions(591ca23c-c75d-445e-a786-9b67689f9cd4).html.
Full textAbstract:
The aim of this work was to assess the ability of the static and dynamic (incorporating the time-course of the inhibitor) prediction models to predict drug-drug interactions (DDIs) using a population-based ADME simulator (Simcyp). This analysis focused on fluconazole, ketoconazole, itraconazole, fluoxetine and fluvoxamine, as CYP inhibitors. The rationale for their selection was an abundance of reported DDI studies, involving a wide range of victim drugs. Preliminary analysis focused on the individual victim drug and inhibitor parameters that are utilised in the DDI prediction models. The victim drug properties included in the DDI prediction models are calculated intrinsically in the Simcyp simulator from in vitro data; these values were compared to estimates obtained by different in vivo methods. Estimations of the fraction metabolised by CYP enzymes were generally consistent with <20% difference between all methods for 15/23 victim drugs. No relationship was observed per CYP enzyme or per inhibitor utilised for phenocopying methods. Estimates of fraction of drug escaping metabolism in the gut were variable across methods with up to 60% coefficient of variation in the case of saquinavir. In vitro assessment of potential liver uptake of the inhibitors was identified for further investigation due to inconsistency in available literature data and sensitivity of the model to this parameter. Extent of liver uptake of selected inhibitors was assessed via comparison of clearance obtained in hepatocytes and microsomes (conventional depletion assay) and values obtained by the conventional depletion and media loss assays in hepatocytes. Clearance was determined at a low concentration (0.1μM) and both rat and human hepatocytes and microsomes were used. The clearance ratios ranged from no difference to >1500 (fluvoxamine from the media loss assay in human hepatocytes). No consistency was observed between methods and human or rat source for any of the inhibitors investigated; therefore, the inclusion of liver uptake into the prediction of DDIs for the current inhibitors was not supported. A database was collated from literature reports of DDIs involving the above named CYP inhibitors (n=97) and used to assess the inclusion of the time-course of inhibition into DDI prediction using the Simcyp simulator. In addition, the impact of active metabolites, dosing time and the ability to predict inter-individual variability in DDI magnitude were investigated using the dynamic prediction model. Simulations comprised of 10 trials with matching population demographics and dosage regimen to the in vivo studies. The predictive utility of the static and dynamic models was assessed relative to the inhibitor or victim drug investigated; both models were employed within Simcyp for consistency in parameters. Use of the dynamic and static models resulted in comparable prediction success, with 67 and 70% of DDIs predicted within two-fold, respectively. Over 60% of strong DDIs (>five-fold AUC increase) were under-predicted by both models, particularly for fluoxetine and fluvoxamine. Incorporation of the itraconazole metabolite into the dynamic model resulted in increased prediction accuracy of strong DDIs (80% within two-fold); no difference was observed for the inclusion of the fluoxetine metabolite. Predicted inter-individual variability in the DDI magnitude was also assessed in healthy, patient and genotyped subjects using a subset of clinical interactions (n=24). Mixed prediction success was observed and the importance of reliable clinical data was highlighted. The differences observed with the dose staggering and the incorporation of active metabolite highlight the importance of these variables in DDI prediction. Finally, the traditional 'two-fold limits' as a measure of the prediction success were reassessed, in particular at AUC ratios approaching one. New limits proposed are applicable for both inhibition and induction DDIs and allow incorporation of the variability in pharmacokinetics of the victim drug when required. DDI predictions were refined using in vitro clearance data for the inhibitors, and assessed using the new predictive measure.
23
Melarkode, Vattekatte Akhila. "3D structures of camelid antibodies : in-silico analyses, prediction and their dynamics." Thesis, La Réunion, 2019. http://www.theses.fr/2019LARE0020.
Full textAbstract:
Les anticorps sont les nouveaux blockbusters au niveau des médicaments. Les anticorps de camélidés, à savoir les domaines VHH, constituent la prochaine génération de traitements et de diagnostics basés sur les anticorps. Pour améliorer nos connaissances sur ces VHH, il est essentiel d'avoir une meilleure vision de la relation séquence, structure et dynamique de ces protéines, et ceci afin d'améliorer leur affinité et leur stabilité. Dans une première étape, une analyse en profondeur des séquences et de la structure des VHHs est effectuée à l'aide d'approches classiques, mais également d'un alphabet structural appelé Blocs Protéiques. Dans une deuxième étape, la relation séquence-structure de ces domaines est évaluée en examinant le lien entre séquences et structures. Dans une troisième étape, la spécificité et la difficulté d’obtenir des modèles structuraux pertinents sont soulignées avec différentes approches, donnant des résultats inattendus. Ensuite, des simulations dynamiques moléculaires à grande échelle ont été effectuées et ont montré une grande diversité dans les comportements de la dynamique des domaines VHH. Enfin, nous terminons notre thèse en énumérant les principales conclusions des chapitres précédents et des perspectives futuresAntibodies are the new blockbusters for drug design. Camelid antibodies, namely VHH domains are projected to be the next generation therapeutics and diagnostics. To improve our knowledge on VHHs, it is essential to have a better view on the sequence, structure and dynamics relationships to improve their affinity and stability. In a first-step a deep analysis of the sequence and the structure of VHHs are done using classical approaches, but also a structural alphabet named Protein Blocks. In a second step, the sequence-structure relationship of these domains is assessed looking at diverse sequence and structures conservations. In a third step, the specificity and difficulty to obtain pertinent structural models are underlined using different approaches, showing unexpected results. Next, the largest scale Molecular dynamic simulations of VHH had been done and shown a large variety in the behaviours of VHH domains dynamics. Finally, the thesis is wrapped up listing significant conclusions from the above chapters and with future perspectives
24
Burt, Howard James. "In Vitro Assessment and Prediction of Time-Dependent CYP3A4 Drug-Drug Interactions." Thesis, University of Manchester, 2008. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.503677.
Full text
25
Yang, Jiansong. "Quantitative prediction of metabolic drug-drug interactions : in vitro - in vivo extrapolation." Thesis, University of Sheffield, 2004. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.422638.
Full text
26
Cochrane, Wolf. "In silico synthesis of analogous lead libraries for drug design by molecular enumeration." Diss., Pretoria : [s.n.], 2007. http://upetd.up.ac.za/thesis/available/etd-04212008-135220/.
Full text
27
Montes, Matthieu. "Développement et applications de méthodes de drug-design et de criblage in silico." Paris 5, 2007. http://www.theses.fr/2007PA05P611.
Full textAbstract:
Les méthodes de criblage in silico basées sur la structure du récepteur sont utilisées pour faciliter la découverte de nouvelles molécules à visée thérapeutique. En utilisant différents outils de docking/scoring, nous avons optimisé un protocole de criblage hiérarchique développé au laboratoire. Ce nouveau protocole a été validé et optimisé pour différentes protéines aux propriétés structurales et physicochimiques très diverses puis appliqué sur deux cibles ayant un rôle déterminant dans différents cancers, la phosphatase à double spécificité CDC25 et le protéasome 20S. En utilisant des chimiothèques filtrées ADME-tox et préparées pour ces projets, nous avons identifié de nouvelles molécules actives sur ces cibles réputées difficiles. Ces inhibiteurs prometteurs ayant une activité in vitro de l’ordre du micromolaire, actifs sur cellules et possédant des squelettes novateurs et optimisables constitueront une base intéressante pour le développement de nouveaux médicaments à visée antiproliférativeVirtual ligand screening methods based on the structure of the receptor are extensively used to facilitate the discovery of lead compounds. Using different docking/scoring packages, we optimized a hierarchical virtual ligand screening protocol developed in our lab. This new protocol has been validated on different targets with different binding site properties. This multi-step hierarchical protocol has been applied to two targets of therapeutic importance, namely, the dual specificity phosphatase CDC25 and the 20S proteasome. Using ADME-tox filtered compound collections processed in our lab, we identified several new active molecules on these two difficult targets. These promising micromolar inhibitors displaying novel and growable scaffolds can lead to new potential drugs for cancer treatment
28
López, del Río Ángela. "Data preprocessing and quality diagnosis in deep learning-based in silico bioactivity prediction." Doctoral thesis, Universitat Politècnica de Catalunya, 2021. http://hdl.handle.net/10803/672385.
Full textAbstract:
Drug discovery is a time and resource consuming process involving the identification of a target and the exploration of suitable drug candidates for it. To streamline drug discovery, computational techniques help identifying molecular candidates with desirable properties by modeling their interactions with the target. These techniques are in constant improvement thanks to the development of algorithms, the increasing computational power and the growth of public molecular databases. Specifically, machine learning approaches provide predictive models on biochemical properties and target-ligand binding activity.
Deep learning is a machine learning approach that automatically extracts multiple levels of representations of the data. Within the last ten years, deep learning has outperformed classical prediction models in most domains, including drug discovery. Common use cases encompass molecular property prediction, de novo compound generation, protein secondary structure prediction and target-compound binding prediction.
However, studies point out the reported performance of deep learning bioactivity prediction models could be a consequence of data bias rather than generalization capability. Efforts are being put in addressing this problem, but it is still present in the state of the art, rewarding novelty over critical assessment. Moreover, the flexibility of deep learning derives in a lack of consensus on how to represent the input spaces, making it difficult to compare models in a common benchmark. Bioactivity data has limited availability because of its associated costs and is often imbalanced, hampering the model learning process. The diagnosis of these problems is not straightforward, since deep learning models are considered black boxes, hindering their adoption as the de facto solution in computer-aided drug discovery.
The present thesis aims to improve deep learning models for computational drug discovery, focusing in the input representation, the data bias control, the data imbalance correction and the model diagnosis.
First, this thesis assesses the effect that different validation strategies have on binding classification models, aiming to find the most realistic performance estimates. The strategy based on clustering molecules to avoid having similar compounds in training and test sets showed to be the most similar to a prospective validation, and thus, more consistent than random cross-validation (over-optimistic) or than an external test set from other database (over-pessimistic).
Second, this thesis focuses on the sequential inputs padding. Padding is necessary to establish a common sequence length by adding zeros to each sequence. These are usually added at the end of the sequence, without formal justification behind it. Here, classical and novel padding strategies were compared in an enzyme classification task. Results showed that the padding position has an effect in the performance of deep learning models, so it should be tuned as an additional hyperparameter.
Third, this thesis studies the effect of data imbalance in protein-compound activity classification models and its mitigation through resampling techniques. The model performance was assessed for different combinations of oversampling the minority class and clustering. Results showed that the proportion of actives predicted by the model was explained by the actual data balance in the test set. Data clustering, followed by data resampling in training and validation sets, stood as the best performing strategy without altering the test set.
To accomplish the three points above, this thesis provides a systematic way to diagnose deep learning models, identifying the factors that govern the model predictions and performance. Specifically, explanatory linear models enabled informed, quantitative decisions regarding input preprocessing. This ultimately leads to more consistent deep learning target-compound binding prediction models.El descubrimiento de fármacos es un proceso costoso en tiempo y recursos. Consiste en la identificación de una diana y la exploración de fármacos candidatos apropiados para ella. Las técnicas computacionales optimizan este proceso, ayudando a identificar las mejores moléculas candidatas mediante el modelado de sus interacciones con la diana. Estas técnicas están en constante mejora gracias al desarrollo de algoritmos, al incremento del poder computacional y al aumento de bases de datos moleculares públicas. Particularmente, el aprendizaje automático proporciona modelos predictivos de distintas propiedades bioquímicas. El deep learning (aprendizaje profundo) es una aproximación del aprendizaje automático basada en las redes neuronales multicapa. Durante los últimos diez años el deep learning ha superado a los modelos predictivos clásicos en la mayoría de dominios, incluído el descubrimiento de fármacos. Algunas de sus aplicaciones son la predicción de propiedades moleculares, la generación de nuevos compuestos, la predicción de la estructura secundaria de proteínas y la predicción de unión entre compuestos y dianas. Sin embargo, algunos estudios apuntan a que el rendimiento reportado por los modelos de deep learning de predicción de unión entre dianas y compuestos podría deberse más al sesgo de los datos que a su capacidad de generalización, dando más peso a la novedad que a la valoración crítica. Además, la flexibilidad del deep learning da pie a una falta de consenso en la representación de sus entradas, dificultando su comparación en un marco común. Los datos de bioactividad tienen una disponibilidad limitada debido a su coste y suelen estar desbalanceados, lo cual puede dificultar el proceso de aprendizaje del modelo. El diagnóstico de estos problemas no es sencillo porque los modelos de deep learning son considerados cajas negras. El objetivo de esta tesis es mejorar los modelos de deep learning para el descubrimiento computacional de fármacos, centrándose en la representación de la entrada, el control del sesgo de los datos, la corrección de su desbalance y el diagnóstico de los modelos. Primero, esta tesis evalúa el efecto de diferentes estrategias de validación en los modelos de clasificación de la unión diana-compuesto para encontrar las estimaciones de rendimiento más realistas. La estrategia basada en el agrupamiento de las moléculas demostró ser la más parecida a una validación prospectiva y por tanto, más consistente que la validación cruzada aleatoria (demasiado optimista) o que un conjunto de test externo proveniente de otra base de datos (demasiado pesimista). Segundo, esta tesis se centra en el relleno de las secuencias de entrada, utilizado para establecer una longitud común de las mismas. Este relleno consiste normalmente en añadir ceros al final de cada secuencia, sin una justificación formal detrás esta decisión. Aquí, se compararon estrategias de relleno novedosas y clásicas en una tarea de clasificación de enzimas. Los resultados mostraron que la posición del relleno tiene un efecto sobre el rendimiento de los modelos de aprendizaje profundo, por lo que se le debería dar más atención. Tercero, esta tesis estudia el efecto del desbalance de los datos en los modelos de clasificación de actividad diana-compuesto y su atenuación mediante técnicas de remuestreo. Se evaluó el rendimiento de un modelo para diferentes combinaciones de sobremuestreo de la clase minoritaria y agrupamiento de las moléculas. Los resultados demostraron que el agrupamiento de los datos, seguido por su remuestreo en los conjuntos de entrenamiento y validación, es la estrategia con mejor rendimiento. Por último, esta tesis proporciona una forma sistemática de diagnosticar modelos de deep learning, identificando los factores que rigen sus predicciones. Estos modelos lineales explicativos permitieron la toma de decisiones informadas y cuantitativas en cada uno
Enginyeria biomèdica
29
Piechota, Przemyslaw. "Development of in silico models for the prediction of toxicity incorporating ADME information." Thesis, Liverpool John Moores University, 2015. http://researchonline.ljmu.ac.uk/4554/.
Full textAbstract:
Drug discovery is a process that requires a significant investment in both time and resources. Although recent developments have reduced the number of drugs failing at the later stages of development due to poor pharmacokinetic and/or toxicokinetic profiles, late stage attrition of drug candidates remains a problem. Additionally, there is a need to reduce animal testing for toxicological risk assessment for ethical and financial reasons. In silico methods offer an alternative that can address these challenges. A variety of computational approaches have been developed in the last two decades, these must be evaluated to ensure confidence in their use. The research presented in this thesis has assessed a range of existing tools for the prediction of toxicity and absorption, distribution, metabolism and elimination (ADME) parameters with an emphasis on absorption and xenobiotic metabolism. These two ADME properties largely determine bioavailability of a drug and, in turn, also influence toxicity. In vitro (Caco-2 cells and the parallel artificial membrane permeation assay) and in silico approaches, such as various druglikeness filters, can be used to estimate human intestinal absorption; a comparison between different methods was performed to identify relative strengths and weaknesses of the approaches. In terms of xenobiotic metabolism it is not only important to predict metabolites correctly, but it is also crucial to identify those compounds that can be biotransformed into species that can covalently bind to biomolecules. Structural alerts are routinely used to screen for such potential reactive metabolites. The balance between sensitivity and specificity of such reactive metabolite alerts has been discussed in the context of correctly predicting reactive metabolites of pharmaceuticals (using data available from DrugBank). Off-target toxicity, exemplified by human Ether-à-go-go-Related Gene (hERG) channel inhibition, was also explored. A number of novel structural alerts for hERG toxicity were developed based on groups of structurally similar compounds. Finally, the importance of predicting potential ecotoxicological effects of drugs was also considered. The utility of zebrafish embryos to distinguish between baseline and excess toxicity was investigated. In evaluating this selection of existing tools, improvements to the methods have been proposed where possible.
30
Aldib, Iyas. "Rational drug design approach of the myeloperoxidase inhibition: From in silico to pharmacological activity." Doctoral thesis, Universite Libre de Bruxelles, 2016. http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/241515.
Full textAbstract:
1. SUMMARYMyeloperoxidase (MPO) which belongs to the peroxidase family, is found in mammalian neutrophils. This heme enzyme contributes to the production of (pseudo)halogenous acid such as HOCl which oxidizes proteins, cell membrane, DNA and RNA causing death for the pathogens. It has an antimicrobial effect due to HOCl secreting inside the phagosomes of the neutrophils, whereas it will be released outside neutrophils causing oxidative damages for the host tissues. Proteins, lipids, lipoproteins, DNA and RNA are potential targets of the MPO resulting in several chronic syndromes. Many researchers have discovered the harmful effects of MPO and its products demonstrating its role in many inflammatory chronic diseases such as: Cardiovascular diseases as in atherosclerosis. MPO contribution in atherosclerosis development has been demonstrated. Neurodegenerative diseases also was related to MPO: such as Alzheimer’s disease (AD), multiple sclerosis (MSc) and Parkinson’s disease The enzyme has been also pointed out in other diseases such as renal disease and cancer.For these reasons, MPO as a target of drug discovery has attracted the attention of many researchers. X-ray 3D structures were resolved for this enzyme, biological activity and mechanism of action were investigated in depth, and many medicinal chemists have investigated and screened for new MPO inhibitors. Indeed, this cumulative work including X-ray data, the role of MPO in different pathologies, MPO inhibitory mechanism of action, screening and various chemical entities that inhibit MPO, provided sufficient elements to start a new drug design and drug discovery process on MPO.The aim of the present study was to apply a rational drug design approach to the myeloperoxidase inhibition: from in silico to pharmacological activity. This includes:─ Conducting high throughput virtual screening in order to find new potential hits to inhibit MPO followed by mechanism of inhibition determination. ─ Selecting one hit and then implementing a whole pharmacomodulation process in order to increase the potency of the inhibition greater than the starting hit and to improve the selectivity.Firstly, a rational drug design process was launched to find new hits using high throughput virtual screening. The chosen database for the screening was ASINEX database published in ZINC.X-ray structure of human peroxidase complexed to cyanide and thiocyanate (PDB 1DNW) was selected to conduct High-Throughput Virtual Screening (HTVS). Three successive protocols with different levels of accuracy in the docking and scoring processes were used starting with HTVS, followed by Standard Precision (SP) and finally with Xtra Precision (XP). The quality of the docking process performed was validated by docking a set of 60 chosen molecules of varying chemical structure and known as MPO inhibitors. From the result of the HTVS conducted on 1,350,000 compounds, the 100 best compounds were selected. Among them, 81 molecules were available for purchase from ASINEX, those compounds were tested with a MPO inhibition assay. Thirty-two compounds (39 %) were active, but only 8 compounds were selected, featuring different chemical structures with IC50 values ranging between 0.46 ± 0.07 and 12 ± 3 μM. Among these molecules, two compounds were the best and considered as hits. One has purinedione structure which is similar with the structure of thioxanthine derivatives (F9, IC50=0.46±0.07μM). The second compound has a hexahydropyrimidine structure (A1, IC50 = 0.5 ± 0.1 μM) The most common interactions found among all 8 docked ligands are the ionic bond with Glu102 and a stacking (shifted or not) with pyrrole ring D of the prosthetic group. Hydrogen bonds with Glu102, Thr100, Gln91, Arg239, or the propionate groups of the heme are also found in several docked geometries of the complexes. Interestingly, interactions with Glu102 and pyrrole ring D of the heme were also seen with fluorotryptamine derivatives and also salicylhydroxamic acid (SHA).For measuring MPO-dependent LDL oxidation, the two best compounds were tested. Compounds A1 and F9 showed good inhibition on MPO-dependent LDL oxidation (62 ± 6, 4.5 ± 0.9, 11 ± 2% and 11 ± 2, 2.6 ± 0.8, 6 ± 4%, respectively).Consequently, in order to determine the mechanism of inhibition transient-state kinetics were further investigated of all the 8 selected compounds.Both new lead compounds (A1 and F9) act as electron donors of both Compound I and Compound II of MPO. The reaction with Compound I was significantly faster (k2 ≫ k3). As a consequence, the enzyme is trapped in the Compound II state. They reversibly inactivated the enzyme blocking the harmful halogenation activity of MPO by transferring it to the MPO peroxidase cycle. In the present study, 8 active and reversible MPO inhibitors were selected. They act as electron donors of the oxidoreductase and efficiently block the halogenation activity with reversible inactivation. Two of the selected compounds have a submicromolar activity and inhibit MPO-dependent LDL oxidation. The high-throughput virtual screening was proved to be a successful tool to find new leads of MPO inhibitors. Conducting HTVS on a large-scale database enabled selection of novel scaffolds of MPO inhibitors never explored before in less time and at less expenses.Finding 8 new different chemical scaffolds through the first step of this drug discovery process led us to choose a new hit, compound A1, which has a hexahydropyrimidine structure, compound F9 was not chosen despite being more active due to its similarity to compounds discovered by AstraZeneca. To conduct pharmacomodulation, a validation of the docking procedure was conducted by comparing the X-ray structures of MPO with 2-(3,5-bistrifluoromethylbenzylamino)-6-oxo-1H-pyrimidine-5- carbohydroxamic acid, HX1, and SHA in the X-ray structures of human MPO in complex with cyanide and thiocyanate (PDB code 1DNW) as well as in complex with HX1 (PDB code 4C1M). Compound A1 was docked into both target structures 1DNW and 4C1M. In both cases, A1 showed almost the same poses.Based on the binding modes of A1, different strategies were developed for the design of derivatives which were mainly focused on the substitution of the aromatic rings A and B, the 2 amino groups and the side chain bridges.Pharmacomodulation was carried out on the hit A1 with different strategies:─ Investigating the role of hydroxyl groups on both aromatic rings─ Shifting the position of the amino groups in the hexahydropyrimidine ring to obtain piperazine derivatives and introduction of fluorine ─ Eliminating of one ring and of an amino group in the hexahydropyrimidine ring leading to piperidine derivatives ─ Opening the hexahydropyrimidine ring while keeping amine function and changing the length of the bridge between this amino group and aromatic ring as well as the impact of substitutions on aromatic rings.─ Hybridization of fluorotryptamine derivatives (effective MPO inhibitors) with hit A1.Based on of the docking experiments, 37 designed compounds were synthesized. The assessment of inhibition of the chlorination activity of MPO was undertaken over the 37 compounds. The hit A1 IC50 = 500 nM. The best compounds inhibiting MPO exhibited the following characteristics:─ One amino group on the bridge between aromatic rings was sufficient for the establishment of binding to Glu102 ─ The presence of three methylene groups between the secondary amine and an aromatic ring improved the inhibition of chlorination and thus decreased the IC50 values. These results showed that the position of the hydroxyl group is important. The distance between the hydrogen bond acceptor (HBA) group of one aromatic ring and the amino group is very important. The docking experiments of bisarylpropylamine derivatives showed ionic and hydrogen bonding interactions between Glu102 and hydroxyl group on aromatic ring linked to the longer side chain.─ Hybridized compounds which carry a fluorotryptamine instead of the phenol ring obtained by hybridization of hit A1 and the potent MPO inhibitors fluorotryptamine derivatives. Actually, compound 38 (which had one aromatic ring and a propyl bridge attached to indole ring) had an IC50 = 54 nM which was 10 times more powerful than the starting hit.The 3 best compounds were tested to examine the transient kinetics. They act as electron donors of the oxidoreductase and efficiently shift MPO from the chlorination cycle to the peroxidase cycle. Due to the similarity of the best compound 38 to serotonin it was tested with the two other best compounds on serotonin transporter (SERT) to examine the selectivity between MPO and SERT.Compound 38 had higher selectivity over MPO but the best selective compound was 28 that contains two aromatic rings carrying one hydroxyl and one fluorine.Electron density maps were conducted to predict the site of oxidation. Results suggested it occurs preferentially at the benzene ring or the indole ring in the best compounds.Determination of redox potentials for the synthesized compounds were tested. Best compounds act as electron donors allowing a one-electron reduction of Compound I.In conclusion, the present study succeeded through rational drug design including structure-based drug design and HTVS to identify new chemical entities for MPO inhibition. Eight compounds were more active than the starting hit A1 with submicromolar inhibition potency. Hybridization and structure based design also gave improvement of selectivity of inhibitors against MPO such as compound 38. Bis-arylalkylamine derivatives are a new group of MPO inhibitors with higher selectivity which could be a new hit for future development.Doctorat en Sciences biomédicales et pharmaceutiques (Pharmacie)
info:eu-repo/semantics/nonPublished
31
Tooley, Adam. "Investigating in vitro methods to aid prediction of in vivo drug-drug interactions." Thesis, University of Sheffield, 2006. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.427244.
Full text
32
Nigsch, Florian. "Computational prediction of molecular properties for drug discovery." Thesis, University of Cambridge, 2009. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.611123.
Full text
33
Wu, Bolin. "PREDICTION OF DRUG INDICATION LIST BY MACHINE LEARNING." Thesis, Uppsala universitet, Statistiska institutionen, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-447232.
Full textAbstract:
The motivation of this thesis originates from the cooperation with Uppsala Monitoring Centre, a WHO collaborating centre for international drug monitoring. The research question is how to give a good summary of the drug indication list. This thesis proposes a regression tree, Random Forests and XGBoost, known as tree-based models to predict the drug indication summary based on its user statistics and pharmaceutical information. Besides, this thesis also compares the aforementioned tree-based models' prediction performance with the baseline models, which are basic linear regression and support vector regression SVR. The analysis shows SVR with RBF kernel and post-pruning tree are the best models to answer the research question.
34
Graham, Helen Sarah. "Prediction of drug distribution in rat and human." Thesis, University of Manchester, 2012. https://www.research.manchester.ac.uk/portal/en/theses/prediction-of-drug-distribution-in-rat-and-human(231d1935-4fde-4b2d-8338-4e74091224f3).html.
Full textAbstract:
Many methods exist in the literature for the prediction of pharmacokinetic parameters which describe drug distribution in rat and human, such as tissue-to-plasma partition coefficients (Kps) and volume of distribution (Vss). However, none of these methods make use of the in vivo information obtained at the early stages of the drug development process in the form of plasma concentration vs. time profiles. The overall aim of the presented study was to improve upon an existing Kp prediction method by making use of the distribution information contained within this experimental data. Chapter 2 shows that Kp values can be successfully obtained experimentally, but that this process is expensive and time-consuming. Chapter 3 compares six Kp prediction methods taken from the literature for their ability to predict the Kp values of 80 drugs. The Rodgers et al. model was found to be the most accurate, with over 77% of predictions within 3-fold of experimental values. This Chapter also discusses the Vss prediction ability of some of these methods, with the Poulin & Theil and Rodgers et al. models shown to be the most accurate predictors for rat Vss and human Vss respectively. Chapter 4 investigates the relationship between muscle Kp and the Kps of all other tissues, to show that experimental muscle Kp can be used as a surrogate from which all other non-adipose Kp values can be predicted. However, the predictions made using this method were shown to be less accurate than predictions made by the Rodgers et al. model for the same dataset of drugs. A relationship was identified between muscle Kp and tumour Kp in rat, suggesting a potential way to predict tumour Kp in the future. In Chapter 5, a novel method is developed whereby Kp predictions made by the Rodgers et al. model are updated using prior information obtained from the in vivo concentration-time profile. These updated values are then used within a physiologically-based pharmacokinetic (PBPK) model and are shown in Chapter 6 to generate improved predictions for other pharmacokinetic parameters such as Vss and clearance in both rat and human. 100% of human Vss predictions made by the most accurate of the novel methods presented here were within 3-fold of experimental values, compared to 68.8% of predictions made by the Rodgers et al. model. The work presented here has highlighted the need for a more accurate method for the prediction of Kp values, and has addressed this need by generating a model which improves upon the most accurate Kp prediction method currently found in the literature. This will lead to an increase in confidence in the use of predicted pharmacokinetic parameters within PBPK modelling.
35
Durdagi, Serdar [Verfasser]. "In silico drug design studies of bioactive cannabinoid and [60]fullerene derivatives / Serdar Durdagi." Berlin : Freie Universität Berlin, 2009. http://d-nb.info/1023624370/34.
Full text
36
Fu, Wai. "In silico prediction of cis-regulatory elements of genes involved in hypoxic-ischaemic insult." Click to view the E-thesis via HKUTO, 2006. http://sunzi.lib.hku.hk/hkuto/record/B36986896.
Full text
37
Fu, Wai, and 符慧. "In silico prediction of cis-regulatory elements of genes involved in hypoxic-ischaemic insult." Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 2006. http://hub.hku.hk/bib/B36986896.
Full text
38
Binti, Mohamad Zobir Siti Zuraidah. "Towards understanding mode-of-action of traditional medicines by using in silico target prediction." Thesis, University of Cambridge, 2018. https://www.repository.cam.ac.uk/handle/1810/270866.
Full textAbstract:
Traditional medicines (TM) have been used for centuries to treat illnesses, but in many cases their modes-of-action (MOAs) remain unclear. Given the increasing data of chemical ingredients of traditional medicines and the availability of large-scale bioactivity data linking chemical structures to activities against protein targets, we are now in a position to propose computational hypotheses for the MOAs using in silico target prediction. The MOAs were established from supporting literature. The in silico target prediction, which is based on the “Molecular Similarity Principle”, was modelled via two models: a Naïve Bayes Classifier and a Random Forest Classifier. Chapter 2 discovered the relationship of 46 traditional Chinese medicine (TCM) therapeutic action subclasses by mapping them into a dendrogram using the predicted targets. Overall, the most frequent top three enriched targets/pathways were immune-related targets such as tyrosine-protein phosphatase non-receptor type 2 (PTPN2) and digestive system such as mineral absorption. Two major protein families, G-protein coupled receptor (GPCR), and protein kinase family contributed to the diversity of the bioactivity space, while digestive system was consistently annotated pathway motif. Chapter 3 compared the chemical and bioactivity space of 97 anti-cancer plants’ compounds of TCM, Ayurveda and Malay traditional medicine. The comparison of the chemical space revealed that benzene, anthraquinone, flavone, sterol, pentacyclic triterpene and cyclohexene were the most frequent scaffolds in those TM. The annotation of the bioactivity space with target classes showed that kinase class was the most significant target class for all groups. From a phylogenetic tree of the anti-cancer plants, only eight pairs of plants were phylogenetically related at either genus, family or order level. Chapter 4 evaluated synergy score of pairwise compound combination of Shexiang Baoxin Pill (SBP), a TCM formulation for myocardial infarction. The score was measured from the topological properties, pathway dissimilarity and mean distance of all the predicted targets of a combination on a representative network of the disease. The method found four synergistic combinations, ginsenoside Rb3 and cholic acid, ginsenoside Rb2 and ginsenoside Rb3, ginsenoside Rb3 and 11-hydroxyprogesterone and ginsenoside Rb2 and ginsenoside Rd agreed with the experimental results. The modulation of androgen receptor, epidermal growth factor and caspases were proposed for the synergistic actions. Altogether, in silico target prediction was able to discover the bioactivity space of different TMs and elucidate the MOA of multiple formulations and two major health concerns: cancer and myocardial infarction. Hence, understanding the MOA of the traditional medicine could be beneficial in providing testable hypotheses to guide towards finding new molecular entities.
39
Rufino, Stephen Duarte. "Analysis, comparison and prediction of protein structure." Thesis, Birkbeck (University of London), 1996. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.243648.
Full text
40
Ye, Zhu Yi Fan. "Deep learning for pharmaceutical formulation prediction." Thesis, University of Macau, 2018. http://umaclib3.umac.mo/record=b3952123.
Full text
41
Ding, Zhenyu. "Molecular prediction of drug response using machine learning methods." Morgantown, W. Va. : [West Virginia University Libraries], 2008. https://eidr.wvu.edu/etd/documentdata.eTD?documentid=5544.
Full textAbstract:
Thesis (M.S.)--West Virginia University, 2008.Title from document title page. Document formatted into pages; contains viii, 65 p. : ill. (some col.). Includes abstract. Includes bibliographical references (p. 63-65).
42
Kilford, Peter John. "In vitro assessment and prediction of drug glucuronidation clearance." Thesis, University of Manchester, 2009. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.499830.
Full text
43
Yaddanapudi, Suryanarayana. "Machine Learning Based Drug-Disease Relationship Prediction and Characterization." University of Cincinnati / OhioLINK, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1565349706029458.
Full text
44
Atreya, Ravi Viswanathan. "Drug target prediction in pancreatic cancer using model organisms." Thesis, The University of Arizona, 2009. http://hdl.handle.net/10150/192261.
Full text
45
Bakal, Mehmet. "Relation Prediction over Biomedical Knowledge Bases for Drug Repositioning." UKnowledge, 2019. https://uknowledge.uky.edu/cs_etds/90.
Full textAbstract:
Identifying new potential treatment options for medical conditions that cause human disease burden is a central task of biomedical research. Since all candidate drugs cannot be tested with animal and clinical trials, in vitro approaches are first attempted to identify promising candidates. Likewise, identifying other essential relations (e.g., causation, prevention) between biomedical entities is also critical to understand biomedical processes. Hence, it is crucial to develop automated relation prediction systems that can yield plausible biomedical relations to expedite the discovery process. In this dissertation, we demonstrate three approaches to predict treatment relations between biomedical entities for the drug repositioning task using existing biomedical knowledge bases. Our approaches can be broadly labeled as link prediction or knowledge base completion in computer science literature. Specifically, first we investigate the predictive power of graph paths connecting entities in the publicly available biomedical knowledge base, SemMedDB (the entities and relations constitute a large knowledge graph as a whole). To that end, we build logistic regression models utilizing semantic graph pattern features extracted from the SemMedDB to predict treatment and causative relations in Unified Medical Language System (UMLS) Metathesaurus. Second, we study matrix and tensor factorization algorithms for predicting drug repositioning pairs in repoDB, a general purpose gold standard database of approved and failed drug–disease indications. The idea here is to predict repoDB pairs by approximating the given input matrix/tensor structure where the value of a cell represents the existence of a relation coming from SemMedDB and UMLS knowledge bases. The essential goal is to predict the test pairs that have a blank cell in the input matrix/tensor based on the shared biomedical context among existing non-blank cells. Our final approach involves graph convolutional neural networks where entities and relation types are embedded in a vector space involving neighborhood information. Basically, we minimize an objective function to guide our model to concept/relation embeddings such that distance scores for positive relation pairs are lower than those for the negative ones. Overall, our results demonstrate that recent link prediction methods applied to automatically curated, and hence imprecise, knowledge bases can nevertheless result in high accuracy drug candidate prediction with appropriate configuration of both the methods and datasets used.
46
Chang, Cheng. "In silico approaches for studying transporter and receptor structure-activity relationships." Connect to this title online, 2005. http://rave.ohiolink.edu/etdc/view?acc%5Fnum=osu1117553995.
Full textAbstract:
Thesis (Ph. D.)--Ohio State University, 2005.Title from first page of PDF file. Document formatted into pages; contains xvii, 271 p.; also includes graphics. Includes bibliographical references (p. 245-269). Available online via OhioLINK's ETD Center
47
Mpangase, Phelelani Thokozani. "Integrating protein annotations for the in silico prioritization of putative drug target proteins in malaria." Diss., University of Pretoria, 2012. http://hdl.handle.net/2263/24715.
Full textAbstract:
Current anti-malarial methods have been effective in reducing the number of malarial cases. However, these methods do not completely block the transmission of the parasite. Research has shown that repeated use of the current anti-malarial drugs, which include artemisinin-based drug combinations, might be toxic to humans. There have also been reports of an emergence of artemisinin-resistant parasites. Finding anti-malarial drugs through the drug discovery process takes a long time and failure results in a great financial loss. The failure of drug discovery projects can be partly attributed to the improper selection of drug targets. There is thus a need for an eff ective way of identifying and validating new potential malaria drug targets for entry into the drug discovery process. The availability of the genome sequences for the Plasmodium parasite, human host and the Anopheles mosquito vector has facilitated post-genomic studies on malaria. Proper utilizationof this data, in combination with computational biology and bioinformatics techniques, could aid in the in silico prioritization of drug targets. This study was aimed at extensively annotating the protein sequences from the Plasmodium parasites, H. sapiens and A. gambiae with data from di fferent online databases in order to create a resource for the prioritization of drug targets in malaria. Essentiality, assay feasibility, resistance, toxicity, structural information and druggability were the main target selection criteria which were used to collect data for protein annotations. The data was used to populate the Discovery resource (http://malport. bi.up.ac.za/) for the in silico prioritization of potential drug targets. A new version of the Discovery system, Discovery 2.0 (http://discovery.bi.up.ac.za/), has been developed using Java. The system contains new and automatically updated data as well as improved functionalities. The new data in Discovery 2.0 includes UniProt accessions, gene ontology annotations from the UniProt-GOA project, pathways from Reactome and Malaria Parasite Metabolic Pathways databases, protein-protein interactions data from. IntAct as well as druggability data from the DrugEBIlity resource hosted by ChEMBL. Users can access the data by searching with a protein identi er, UniProt accession, protein name or through the advanced search which lets users filter protein sequences based on different protein properties. The results are organized in a tabbed environment, with each tab displaying different protein annotation data. A sample investigation using a previously proposed malarial target, S-adenosyl-Lhomocysteine hydrolase, was carried out to demonstrate the diff erent categories of data available in Discovery 2.0 as well as to test if the available data is su fficient for assessment and prioritization of drug targets. The study showed that using the annotation data in Discovery 2.0, a protein can be assessed, in a species comparative manner, on the potential of being a drug target based on the selection criteria mentioned here. However, supporting data from literature is also needed to further validate the findings.Dissertation (MSc)--University of Pretoria, 2012.
Biochemistry
unrestricted
48
Turpeinen, M. (Miia). "Cytochrome P450 enzymes—in vitro, in vivo, and in silico studies." Doctoral thesis, University of Oulu, 2006. http://urn.fi/urn:isbn:9514282205.
Full textAbstract:
Abstract
Metabolism is a major determinant of the pharmacokinetic properties of most drugs and is often behind bioavailability problems, drug-drug interactions, and metabolic idiosyncrasies. Cytochrome P450 (CYP) enzymes are a superfamily of microsomal hemoproteins catalysing the metabolic reactions of several exogenous compounds. The majority of crucial steps within drug metabolism are in connection with CYP enzymes.
In the present study, in vivo, in vitro, and in silico approaches were applied and characterised to evaluate the effects of chemical entities on CYP-mediated metabolism. CYP2B6 was used as a target enzyme for these studies.
For evaluation of the CYP inhibition potential of new chemical entities, a novel in vitro test system utilising the n-in-one approach was developed. This method proved to be robust and applicable to screening purposes. Validation of the n-in-one assay was done by comparing its performance to commonly used in vitro techniques using six structurally diverse drugs. All assay types yield remarkably similar results with the majority of the CYP forms tested.
Several chemicals were screened in vitro and in silico in order to find potent and selective chemical inhibitors for CYP2B6. Ticlopidine, thioTEPA and 4-(4-chlorobenzylpyridine) were found to be highly effective inhibitors of CYP2B6. The selectivity of thioTEPA proved to be very high, whereas ticlopidine and 4-(4-chlorobenzylpyridine) also inhibited other CYPs. At a concentration level of 1 μM for ticlopidine and 0.1 μM for 4-(4-chlorobenzylpyridine), the inhibitory effect towards other CYPs was negligible.
Due to wide clinical use and relevance, clopidogrel and ticlopidine were selected for further in vivo interaction studies. Both clopidogrel and ticlopidine significantly inhibited the CYP2B6-catalysed bupropion hydroxylation and patients receiving either clopidogrel or ticlopidine are likely to need dose adjustments when treated with drugs primarily metabolised by CYP2B6. The effect of impaired kidney function on CYP2B6 activity and on bupropion pharmacokinetics was also explored. In patients with kidney disease, the bupropion AUC and Cmax were significantly higher and the apparent oral clearance of bupropion was notably lower compared to healthy controls.
The present results indicate that the in silico and in vitro methods used are helpful in predicting in vivo drug-drug interactions. The effective utilisation of these models in the early phases of drug discovery could therefore help to target the in vivo studies and to eliminate metabolically unfavourable drug candidates.
49
Salentin, Sebastian. "In Silico Identification of Novel Cancer Drugs with 3D Interaction Profiling." Doctoral thesis, Saechsische Landesbibliothek- Staats- und Universitaetsbibliothek Dresden, 2018. http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa-226435.
Full textAbstract:
Cancer is a leading cause of death worldwide. Development of new cancer drugs is increasingly costly and time-consuming. By exploiting massive amounts of biological data, computational repositioning proposes new uses for old drugs to reduce these development hurdles. A promising approach is the systematic analysis of structural data for identification of shared binding pockets and modes of action.
In this thesis, I developed the Protein-Ligand Interaction Profiler (PLIP), which characterizes and indexes protein-ligand interactions to enable comparative analyses and searching in all available structures. Following, I applied PLIP to identify new treatment options in cancer: the heat shock protein Hsp27 confers resistance to drugs in cancer cells and is therefore an attractive target with a postulated drug binding site. Starting from Hsp27, I used PLIP to define an interaction profile to screen all structures from the Protein Data Bank (PDB). The top prediction was experimentally validated in vitro. It inhibits Hsp27 and significantly reduces resistance of multiple myeloma cells against the chemotherapeutic agent bortezomib.
Besides computational repositioning, PLIP is used in docking, binding mode analysis, quantification of interactions and many other applications as evidenced by over 12,000 users so far. PLIP is provided to the community online and as open source.
50
Hobbs, Michael. "The interplay of multiple ADME mechanisms : prediction of hepatic drug-drug interactions from in vitro." Thesis, University of Strathclyde, 2016. http://digitool.lib.strath.ac.uk:80/R/?func=dbin-jump-full&object_id=27533.
Full textAbstract:
Rosuvastatin has poor passive membrane permeability and its uptake into the liver is mediated predominately by the transporter, Organic Anion Transporting Polypeptide 1B1 (OATP1B1). Cyclosporin A (CsA) is a potent inhibitor of a range of transporters, including OATP1B1 and clinical drug-drug interactions (DDI) have been reported with rosuvastatin. The aim of this study was to determine the uptake kinetics of rosuvastatin in human hepatocytes using a mechanistic model and to determine the inhibitory effect of CsA upon those kinetics. These data may then allow the extrapolation of in vitro to in vivo kinetics and provide an understanding of the interplay between different disposition mechanisms, with particular regard to the potential to predict DDI.This study was divided into method development and experimental phases. In the development phase, paediatric hepatocytes from a single donor were used to develop the methods. The uptake parameters (Vmax, Km,u, Pdiff,u, Fucell and CLuptake) using estradiol-17β-D-glucuronide (EG) and rosuvastatin were determined using a mechanistic two-compartment model developed by Menochet et al (2012a). The uptake of rosuvastatin was also determined using sodium free media, which prevents the efficient functioning of the uptake transporter Sodium Taurocholate Dependent Transporter (NTCP), another transporter thought to contribute to the uptake of rosuvastatin. Inhibition parameters (IC50) of the uptake of EG and rosuvastatin by CsA and rosuvastatin and rifampicin were determined. The uptake kinetic parameters of EG and rosuvastatin in the paediatric human hepatocytes were in agreement with the quoted literature values for adult human hepatocytes. The hepatocytes were robust enough to be used for method development and to plan for the future studies. The IC50 values for EG and rosuvastatin as the probe substrates using CsA and rifampicin as inhibitors were in agreement with quoted literature values and suggested a predominate role for OATP1B1 in EG and rosuvastatin uptake. In these paediatric human hepatocytes NTCP did not appear to play a role in the uptake of rosuvastatin. The paediatric human hepatocyte data were used to help define and refine the studies conducted in the experimental phase of the study. The same uptake parameters for rosuvastatin were determined in human hepatocytes from three adult donors using the mechanistic two-compartment model. The time of the incubation was extended to 60 minutes to ensure that steady state kinetics were reached. Inhibition of the uptake of rosuvastatin was determined with co- and pre-incubation of CsA and its main metabolite, AM1. The time-dependent nature of the inhibitors have been studied by ourselves, but not in human hepatocytes [Gertz et al 2012]. There did not appear to be an effect of co- versus pre-incubation of the inhibitors. The mechanistic two-compartment model was also used to determine the uptake parameters for a GSK compound, namely GSK2879552, using the same adult human hepatocytes from three adult donors. However, the hepatic uptake clearance (CLuptake) values suggested that active saturable uptake of GSK2879552 was not evident. This may explain the high variability observed between the three donors for the uptake parameters (Vmax, Km,u, Pdiff,u and Fucell) and the high coefficient of variation observed about each parameter. These data provided a useful learning with regards to understanding the limitations of the model.