Tesi sul tema "Computer simulations"

In this dissertation I have created and applied a parametric model for bulk carbonate materials. The new empirical model for carbonates is stable for a wide range of carbonate structures and reproduces experimental results with reasonable accuracy. To study the surface of calcite the ab initio code SIESTA has been used. New implementation has been introduced into the SIESTA code to allow the calculation of effective charges using the modern theory of polarisation. Using these charges the calculation of the long range electrostatic effects, which are removed by the zero electric field boundary conditions, have been introduced into the phonon methodology, reproducing the LO-TO splitting within the calculated phonon modes near the F-point. Furthermore the effective charges have been used in the calculation of the infrared intensity for each phonon mode. The SIESTA implementation of DFT relies upon the evaluation of electron density on a real-space grid. Such discretization of the real-space integrals introduces an oscillatory error in the energy and forces, with the periodicity of the real-space grid. A method for reducing this error has been introduced. The SIESTA code with the new methodology has been used to study bulk calcite, {211} calcite surface and the interaction of water with the {211} sur­face. The structure and phonon frequencies for the bulk match well with experimental values. The {211} surface has been calculated showing the response of the crystal in both distortion of the ion position and the electronic configuration. Surface relaxations and phonon frequencies show no symmetry breaking reconstruction of the calcite {211} surface. Calculation of the interaction of water molecules with the {211} surface predicts the optimum position for water on the surface.

Peptide bond formation and translational termination on the ribosome have been simulated by molecular mechanics, free energy perturbation, empirical valence bond (MD/FEP/EVB) and automated docking methods. Recent X-ray crystallographic data is used here to calculate the entire free energy surface for the system complete with substrates, ribosomal groups, solvent molecules and ions. A reaction mechanism for peptide bond formation emerges that is found to be catalyzed by the ribosome, in agreement with kinetic data and activation entropy measurements. The results show a water mediated network of hydrogen bonds, capable of reducing the reorganization energy during peptidyl transfer. The predicted hydrogen bonds and the structure of the active site were later confirmed by new X-ray structures with proper transition states analogs.

Elongation termination on the ribosome is triggered by binding of a release factor (RF) protein followed by rapid release of the nascent peptide. The structure of the RF, bound to the ribosomal peptidyl transfer center (PTC), has not been resolved in atomic detail. Nor is the mechanism known, by which the hydrolysis proceeds. Using automated docking of a hepta-peptide RF fragment, containing the highly conserved GGQ motif, we identified a conformation capable of catalyzing peptide hydrolysis. The MD/FEP/EVB calculations also reproduce the slow spontaneous release when RF is absent, and rationalize available mutational data. The network of hydrogen bonds, the active site structure, and the reaction mechanism are found to be very similar for both peptidyl transfer and termination.

New structural data, placing a ribosomal protein (L27) in the PTC, motivated additional MD/FEP/EVB simulations to determine the effect of this protein on peptidyl transfer. The simulations predict that the protein N terminus interacts with the A-site substrate in a way that promotes binding. The catalytic effect of L27 in the ribosome, however, is shown to be marginal and it therefore seems valid to view the PTC as a ribozyme. Simulations with the model substrate puromycin (Pmn) predicts that protonation of the N terminus can reduce the rate of peptidyl transfer. This could explain the different pH-rate profiles measured for Pmn, compared to other substrates.

It has been recognized for some time that persons with severe physical disabilities could benefit greatly if they had a personal robot under their control. Such kind of robots has been available for several years, but the acceptence by disabled persons has been very slow. One of the reasons is a concern for safety. If a robot arm is strong enough to bring food to a person's mouth, then it could do severe personal injury to the eyes or teeth in the event of an electronic or mechanical failure. In this thesis, a new approach of investigating a personal robot will be presented. A three-dimensional computer simulation of a personal robot will be described. It is completely under the control of the user. This simulation system has many advantages over testing an actual model, since it enables almost any type of personal robot and any type of control strategy to be investigated. This system can be used by researchers to investigate control algorithms for a personal robot. It can also be used by disabled persons so that they can get familiar with the robot as well as the control strategy before they use or purchase the actual model. Currently, this simulation is used to investigate a control strategy called "Modified Extended Physiological Proprioception".

Computer simulations of biological systems provide novel data while both supporting and challenging traditional experimental methods. However, continued innovation is required to ensure that these technologies are able to work with increasingly complex systems. Coarse–grained approximations of protein structure have been studied using a lattice model designed to find low–energy conformations. A hydrogen–bonding term has been introduced. The ability to form β–sheet has been demonstrated, and the intricacies of reproducing the more complex α–helix on a lattice have been considered. An alternative strategy, that of better utilising computing power through the technique of milestoning, has shown good agreement with previous experimental and computational work. The increased efficiency allows significantly less extreme simulation conditions to be applied than those used in alternative simulation methods, and allows more simulation repeats. Finally, the principles of Least Action Dynamics have been employed to combine the two approaches described above. By splitting a simulation trajectory into a number of smaller components, and using the lattice model to optimise the path from a start structure to an end structure, it has been possible to efficiently generate dynamical information using an alternative method to traditional molecular dynamics.

We present here the results of a series of molecular dynamics (MD) simulations of systems of soft repulsive tapered particles. Essentially, the objective of the project was to investigate the effect that changing the degree of taper of these particles has on the collective behaviour of the system. The particle shapes were modelled using the parameterised Gaussian overlap (PHGO) contact function, which had previously been used in Monte Carlo (MC) studies of systems of hard particles with the same range of shapes [Phys. Rev. E 68,021708 (2003)]. The work carried out falls into three main categories. Firstly we calculated the splay and bend reduced flexoelectric coefficients, e₁₁ and e₃₃, for a number of systems in the nematic phase. This was done using the linear response approach developed by Nemtsov and Osipov [Kristallografiya 31 2,213-218 (1986)]. The values of e₁₁ measured for the tapered systems studied were all positive and of the order of +0.1, whilst the e₃₃ values were of a similar magnitude but negative in all cases. These numbers correspond to values of the order of pCm^-1, which is consistent with typical values measured experimentally for the flexoelectric coefficients. The reduced coefficients for systems of uniaxial particles were also calculated and found to be approximately zero, as they should be for particles with this type of symmetry. The second major theme in the project was the mapping out of the shape-density phase diagram, through both compression and decompression sequences, for tapered particles having a constant length to breadth ratio of 3 but different degrees of tapering, ranging from an extreme tear-drop shape to the uniaxial Gaussian ellipsoid. The results of our MD simulations broadly agreed with those obtained by the MC route [Phys. Rev. E 68,021708 (2003)]. Isotropic, nematic, smectic and ordered solid phases were clearly identified. In addition a so-called `curvy-bilayer' (CB) phase was observed, which locally possessed similar order to the smectics but did not exhibit any clear long range order. The structure of the CB phase was investigated further and found to be a type of bicontinuous cubic phase, specifically the Ia3d or gyroid (G) as it is also known -a phase never before obtained from this type of simulation. Characterisation of the I-G transition was undertaken, which indicated that the gyroid freely selfassembled from precursors present in the isotropic fluid. The Sm-G transition was also characterised and found to take place via the formation, firstly of stalks joining two adjacent smectic bilayers and then subsequently pores which bisect these bilayers and initiate the emergence of the gyroid morphology.

This study outlines the design, implementation, and testing of the General Control Model as applied to the Future Theater-Level Model (FTLM) for the control of Joint and Allied Forces for all operational sides. The study develops a notion of battlefield control and describes the characteristics necessary to represent this notion of control in a computer simulation. Central to the implementation of the General Control Model is the robust capability for the user-analyst to describe any control relationship of research interest and to do so without having to alter the programming code. The user-analyst is provided the capability to determine the cause and effect relationship of different control representations in a simulation. A full description of the model is complimented by an explanation of the implementation to facilitate the use of the General Control Model. A discussion of the initial test results leads to a more rigorous test which confirms the intended behavior of the General Control Model in FTLM. Lastly, recommendations for future improvements to the General Control Model and FTLM are outlined to assist future research endeavors.

Molecular modeling has come a long way during the past decades and in the current thesis modeling of biological membranes is the focus. The main method of choice has been classical Molecular Dynamics simulations and for this technique a model Hamiltonian, or force field (FF), has been developed for lipids to be used for biological membranes. Further, ways of more accurately simulate the interactions between solutes and membranes have been investigated. A FF coined Slipids was developed and validated against a range of experimental data (Papers I-III). Several structural properties such as area per lipid, scattering form factors and NMR order parameters obtained from the simulations are in good agreement with available experimental data. Further, the compatibility of Slipids with amino acid FFs was proven. This, together with the wide range of lipids that can be studied, makes Slipids an ideal candidate for large-scale studies of biologically relevant systems. A solute's electron distribution is changed as it is transferred from water to a bilayer, a phenomena that cannot be fully captured with fixed-charge FFs.  In Paper IV we propose a scheme of implicitly including these effects with fixed-charge FFs in order to more realistically model water-membrane partitioning. The results are in good agreement with experiments in terms of free energies and further the differences between using this scheme and the more traditional approach were highlighted. The free energy landscape (FEL) of solutes embedded in a model membrane is explored in Paper V. This was done using biased sampling methods with a reaction coordinate that included intramolecular degrees of freedom (DoF). These DoFs were identified in different bulk liquids and then used in studies with bilayers. The FELs describe the conformational changes necessary for the system to follow the lowest free energy path. Besides this, the pitfalls of using a one-dimensional reaction coordinate are highlighted.

Molecular simulations performed on modern computers provide a powerful tool for the investigation of both static and dynamic properties of liquid crystals. In this thesis several properties of liquid crystal mesogens have been investigated using state-of-the-art Monte Carlo (MC) and molecular dynamics (MD) simulation techniques. The helical twisting power, βm, determines the pitch of the chiral nematic phase produced when a nematic liquid crystal is doped with a low concentration of chiral solute molecules. A new simulation technique that allows the prediction of both the sign and the magnitude of βm is described. The method employs fully atomistic MC simulations of a chiral dopant molecule in the presence of a twisted nematic solvent composed of Gay-Berne particles. Eighteen different chiral dopant molecules were examined and in all cases the results were in good agreement with existing experimental data. The Kirkwood correlation factor, g(_1), has been evaluated for the molecules PCH5, PCH5-C1, me5NF and GGP5C1 using MD simulations in the pre-transitional region of the isotropic phase. The calculations employed an all-atom force field, which was developed specifically for liquid crystal molecules. PCH5 and meSNF were seen to favour anti-parallel dipole alignment whereas, PCH5-C1 and GGP5C1 preferred a parallel arrangement of the molecular dipoles. With the exception of GGP5C1, the simulations gave g(_1) values that were in accordance with existing experimental dielectric measurements. Detailed analysis of the MD trajectories showed that certain molecular pair configurations were preferred in the bulk and indicated which molecular groups were responsible for the stabilization of these configurations. Equilibrium molecular dynamics simulations were carried out in order to evaluate the rotational viscosity coefficient, γ(_1), for a Gay-Berne mesogen using two independent analysis techniques. The methods gave consistent results, which were comparable to experimental data for real mesogens of similar shape and size.

This dissertation considers a number of numerical simulations of crushable grains. It begins by describing the algorithms used in a 2-dimensional model in which grains crush according to a set of statistical rules, resulting in the emergence of fractal distributions of particle sizes and linear compression plots. In this simulation, grains were represented by closely packed triangles that fractured probabilistically on the basis of their size, number of neighbours and an overall macroscopic stress parameter. These algorithms allowed the location and identity of neighbouring elements to be determined efficiently for large numbers of elements, over size ranges of several orders of magnitude. In the initial simulation voids were not explicitly represented and these were added into a second simulation in which elements fractured according to a similar set of rules but were permitted to move according to a set of simple kinematic rules. The differences in the results obtained from each of these simulations and their limitations of are discussed. A discrete element package PFC3D is then used to model individual particles in simple crushing tests. 3-dimensional particles are constructed from smaller, bonded elements and random variation is introduced into the construction of the particles so that the distribution of strengths of batches of particles can be reasonably represented by a Weibull distribution. The thesis ends with some preliminary stress paths conducted on assemblies of such particles. At a superficial level contours of broken bonds bear a resemblance to the ,shapes of curves predicted by the Cambridge plasticity model Cam Clay. These simulations are therefore a promising point of departure for future investigation of the physical meaning of parameters describing these more complex aspects of soil behaviour.

In this thesis we studied nematic liquid crystals using molecular dynamics simulations based on the coarse grained Gay-Berne potential. The elastic and dynamical properties of the nematic bulk were calculated and the impact of the system size and simulation run time were investigated showing that both have to be considered carefully. For the bend fluctuations we observed propagating modes for the director and velocity components. This contradicts statements found in the literature that assume these modes are overdamped. We derive from nematodynamics that this assumption may not be valid for all systems and hence we argue that propagating modes may be observed in experiments. Furthermore we studied defect structures forming due to nanoparticle inclusions in nematics. Depending on the particle size and the surface anchoring three different defect types were observed: Saturn ring, surface-ring and boojum defects. The satellite defect was found to be unstable for the particle sizes studied here, which is in agreement with theoretical predictions. For two nanoparticles in close proximity entangled defects formed, similar to experimental observations for micron sized particles. We explain the three-ring structure, which was observed in other molecular simulations, as a superposition of the different entangled states. Finally we calculated the line tension and viscous drag of a single disclination line of strength -1/2. Nanoparticles placed in close proximity of the single disclination experienced highly non-linear attractive forces. Once the particle ‘touches’ the disclination it remained connected for the entire simulation. In addition we have shown that the presence of a single disclination has a significant impact on inter-nanoparticle interactions.

Thesis (Ph. D.)--Materials Science and Engineering, Georgia Institute of Technology, 2008.
Committee Chair: Dr. Arun Gokhale; Committee Member: Dr. Hamid Garmestani; Committee Member: Dr. Karl Jacob; Committee Member: Dr. Meilin Liu; Committee Member: Dr. Steve Johnson.

The important properties of biological membranes such as elasticity and structure, and their interaction with nanoparticles is of great importance to nanomedicine applications such as drug delivery and gene therapy. This thesis reports on studies carried out using molecular dynamics (MD) simulations to investigate the physics and chemistry of POPC lipid bilayer membranes and their interactions with ions and nanoparticles. In this study two techniques were employed; all atomic (AA) and coarse grained (CG) MD simulations. In the first part of our investigations, the elastic properties of a 1-palmitoyl-2-oleoyl-snglycero- 3-phosphocholine (POPC) membrane with missing leaflets as well as a defect-free membrane were determined. The calculated bulk moduli compare well with the results of other researchers. Most interestingly, we established that the removal of a whole leaflet or half a leaflet does significantly affect the elastic properties. In studying of diffusivity of Na and Cl ions as well as of water molecules across a POPC membrane, we constructed systems to model the flow of ions across a membrane separating two different aqueous solutions. We were able to show that in order to force diffusion across the membrane, it was necessary to introduce an imbalance of positively and negatively charged ions on either side of the membrane. We have been able to confirm that the diffusion process takes place by the creation of a pore in the membrane. The diffusion coefficients of the ions have been determined from the mean square displacements (MSD) of the particles as a function of time. We found that both the Na and Cl ions diffuse rapidly through the pore with diffusion coefficients ten times larger than in water. Also, we observed that although the Na ions are the first to begin the permeation process due to the lower potential barrier it experiences, the Cl ions complete the permeation across the barrier more quickly due to their faster diffusion rates. AAMD simulations were carried out to determine the adsorption sites of neutral and charged gold nanoparticles on the surface of the POPC membrane in order to understand the first step of translocation process. We have found that the adsorption of a neutral gold nanoparticle is more likely than that of a charged nanoparticle. We were also able to demonstrate the partial penetration of a neutral nanoparticle through the surface of the POPC membrane. CGMD was used to study the stability of a carbon nanotube (CNT) inside the POPC lipid bilayer membrane. We found that the CNT was indeed stable inside the POPC membrane suggesting that such nanotubes could be used as a targeted drug delivery.

This thesis aims to develop a computer simulation of the process that occurs when one displaces a viscous fluid such as oil by a less viscous one such as water in a porous medium. Chapter 1 outlines the problem and explains why a computer simulation rather than analytical treatment is necessary for the problem. Previous computer simulations of the problem are reviewed and their respective advantages and disadvantages are considered. Chapter 2 introduces the concept of 'simulated annealing', a stochastic computational technique for solving minimisation problems with many variables and this technique is used to make a crude model of the displacement problem. The results from this are considered and the reasons for the model's failure to adequately solve the problem are discussed. In chapter 3, simulated annealing is applied to the simpler problem of the travelling salesman where one has to find the shortest route around a collection of points. The aim of this chapter is to try and find an optimum simulated annealing schedule to minimise the computer time needed to achieve a satisfactory solution. This is successfully accomplished for this particular problem by fitting the relaxation time of the system as a function of temperature to an Arrhenhius type law. But this optimisation is problem specific and it is concluded that the complicated nature of the oil displacement problem effectively precludes treatment by annealing. In chapter 4 a stochastic micro model is developed in which a pressure gradient across the system forces water into oil bearing pores. The pores have varying sizes which represent sizes which represent the varying permeability in a porous medium. A modified Gauss Seidel method is used to solve for the pressure field and an analytic expression for the saturation update is developed. The final chapter, chapter 5, develops the above model further and in particular develops a scheme whereby conservation of fluid is guaranteed. The profiles of the fingering of the water into the oil are studied and it is found that their interface fractal dimension varies monotonically with viscosity ratio.

Self-assembly of colloidal particles into ordered structures is hailed as the preferred route to production of functional devices on the nanometre and micron length scales. The shape of a colloidal particle is one of the most influential factors determining the type of ordered structure that is assembled. Thus this thesis is devoted to understanding the role of particle shape on phase behaviour of colloidal systems. The effect of particle shape is isolated by using computer simulations to model particles as hard, anisotropic bodies which interact via purely repulsive interactions. Two particle models are studied which are representative of real colloids: non-convex wireframe polyhedra, and convex spherical caps. This thesis investigates the densest packings of several wireframe polyhedra. By comparing packings of six distinct polyhedra some general conclusions are drawn regarding the effects of rounded polyhedra edges, and a new shape descriptor is given which can suggest whether a wireframe polyhedron is likely to form new interpenetrating crystal structures. Wireframe cubes were studied in more detail, where the full phase behaviour was mapped out. A curious phenomenon was found whereby crystals formed by cubic wireframes exhibit plastic fluctuations. This unusual behaviour, if reproduced experimentally, may lead to useful optical properties. A systematic study of spherical caps demonstrates the effect of shape on collective behaviour as the particle model interpolates between a sphere and a thin platelet. Purely repulsive interactions are responsible for a range of different crystal structures, but the nucleation of these structures is challenging due to slow dynamics. Furthermore, there are often many ways for a spherical cap to pack in a given volume, which leads to multiple metastable states. The self-assembly of spherical caps was directed by sedimentation on a solid template which resulted in increased nucleation rates and more stable crystals. However, there is still a lack of control over the exact crystal structure due to the degeneracy in ways to pack.

Computer simulations have become a vital tool in modern science. The ability to reliably move beyond the capabilities of experiment has allowed great insights into the nature of matter. To enable the study of a wide range of systems and properties a plethora of simulation techniques have been developed and refined, allowing many aspects of complex systems to be demystified. I have used a range of these to study a variety of systems, utilising the latest technology in high performance computing (HPC) and novel, nanoscale models. Monte Carlo (MC) simulation is a commonly used method to study the properties of system using statistical mechanics and I have made use of it in published work [1] to study the properties of ferrogels in homogeneous magnetic fields using a simple microscopic model. The main phenomena of interest concern the anisotropy and enhancement of the elastic moduli that result from applying uniform magnetic fields before and after the magnetic grains are locked in to the polymer-gel matrix by cross-linking reactions. The positional organization of the magnetic grains is influenced by the application of a magnetic field during gel formation, leading to a pronounced anisotropy in the mechanical response of the ferrogel to an applied magnetic field. In particular, the elastic moduli can be enhanced to different degrees depending on the mutual orientation of the fields during and after ferrogel formation. Previously, no microscopic models have been produced to shed light on this effect and the main purpose of the work presented here is to illuminate the microscopic behaviour. The model represents ferrogels by ensembles of dipolar spheres dispersed in elastic matrices. Experimental trends are shown to be reflected accurately in the simulations of the microscopic model while shedding light on the microscopic mechanism causing these effects. These mechanisms are shown to be related to the behaviour of the dipoles during the production of the gels and caused by the chaining of dipoles in magnetic fields. Finally, simple relationships between the elastic moduli and the magnetization are proposed. If supplemented by the magnetization curve, these relationships yield the dependencies of the elastic moduli on the applied magnetic field, which are often measured directly in experiments. While MC simulations are useful for statistical studies, it can be difficult to use them to gather information about the dynamics of a system. In this case, Molecular Dynamics (MD) is more widely used. MD generally utilises the classical equations of motion to simulate the evolution of a system. For large systems, which are often of interest, and multi-species polymers, the required computer power still poses a challenge and requires the use of HPC techniques. The most recent development in HPC is the use of Graphical Processing Units (GPU) for the fast solution of data parallel problems. In further published work [2], I have used a bespoke MD code utilising GPU acceleration in order to simulate large systems of block copolymers(BC) in solvent over long timescales. I have studied thin films of BC solutions drying on a flat, smooth surface which requires long timescales due to the ’slow’ nature of the process. BC’s display interesting self-organisation behaviour in bulk solution and near surfaces and have a wide range of potential applications from semi-conductors to self-constructing fabrics. Previous studies have shown some unusual behaviour of PI-PEO diblock co-polymers adsorbing to a freshly cleaved mica surface. These AFM studies showed polymers increasing in height over time and proposed the change of affinity of mica to water and the loss of water layers on the surface as a driver for this change. The MD simulation aimed to illuminate the process involved in this phenomena. The process of evaporation of water layers from a surface was successfully simulated and gave a good indication that the process of solvent evaporation from the surface and the ingress of solvent beneath the adsorbed polymer caused the increase in height seen in experiment.

Carbohydrate molecules are essential parts of living cells. They are used as energy storage and signal substances, and they can be found incorporated in the cell membranes as attachments to glycoproteins and glycolipids, but also as free molecules. In this thesis the effect of carbohydrate molecules on phospholipid model membranes have been investigated by the means of Molecular Dynamics (MD) computer simulations. The most abundant glycolipid in nature is the non-bilayer forming monogalactosyldiacylglycerol (MGDG). It is known to be important for the membrane stacking typical for the thylakoid membranes in plants, and has also been found essential for processes related to photosynthesis. In Paper I, MD simulations were used to characterize structural and dynamical changes in a lipid bilayer when MGDG is present. The simulations were validated by direct comparisons between dipolar couplings calculated from the MD trajectories, and those determined from NMR experiments on similar systems. We could show that most structural changes of the bilayer were a consequence of lipid packing and the molecular shape of MGDG. In certain plants and organisms, the enrichment of small sugars such as sucrose and trehalose close to the membrane interfaces, are known to be one of the strategies to survive freezing and dehydration. The cryoprotecting abilities of these sugar molecules are long known, but the mechanisms at the molecular level are still debated. In Papers II–IV, the interactions of trehalose with a lipid bilayer were investigated. Calculations of structural and dynamical properties, together with free energy calculations, were used to characterize the effect of trehalose on bilayer properties. We could show that the binding of trehalose to the lipid bilayer follows a simple two state binding model, in agreement with recent experimental investigations, and confirm some of the proposed hypotheses for membrane–sugar interactions. The simulations were validated by dipolar couplings from our NMR investigations of TRH in a dilute liquid crystal (bicelles). Furthermore, the assumption about molecular structure being equal in the ordered and isotropic phases was tested and verified. This assumption is central for the interpretation of experimentally determined dipolar couplings in weakly ordered systems. In addition, a coarse grain model was used to tackle some of the problems with slow dynamics that were encountered for trehalose in interaction with the bilayer. It was found that further developments of the interaction models are needed to properly describe the membrane–sugar interactions. Lastly, from investigations of trehalose curvature sensing, we concluded that it preferably interacts in bilayer regions with high negative curvature.

At the time of the doctoral defense, the following paper was unpublished and had a status as follows: Paper 4: Manuscript.

本文データは平成22年度国立国会図書館の学位論文(博士)のデジタル化実施により作成された画像ファイルを基にpdf変換したものである
Kyoto University (京都大学)
0048
新制・課程博士
博士(情報学)
甲第8488号
情博第14号
新制||情||2(附属図書館)
UT51-2000-F392
京都大学大学院情報学研究科通信情報システム専攻
(主査)教授 松本 紘, 教授 橋本 弘蔵, 教授 大村 善治
学位規則第4条第1項該当

Magnetisation reversal is a process that is of paramount technological importance, as well as fascinating scientists in the field. Despite its importance, only recently have scientists begun to probe the limits of what can be achieved. Driven by consumer desire for compact devices using large scale storage, the size and time limit of magnetisation reversal have begun to be unravelled by exciting experiments using large scale national facilities and state of the art modelling. In the work presented here we develop a model of a ferrimagnetic material parameterised on experimental observations. The key magnetic features are shown to agree well with experimental measurements and provide a basis for more complex calculations. Using time and element resolved X-ray magnetic circular dichroism experiments, model calculations of laser induced magnetisation dynamics are compared to experimental measurements. We present results of switching on the sub-picosecond timescale and conclude that it is possible to reverse magnetisation using heat alone. The conclusion that magnetisation can be reversed without a directional stimulus is scientifically intriguing and never before predicted. Confirmation of such a reversal mechanism using heat alone is verified experimentally, comparing well with the model predictions.

In order to further understand the behaviour of magnetic recording media the relationship between the microscopic and the macroscopic behaviour must be more precisely understood. Computer simulations are an invaluable tool in the development and testing of models of these media. Presented here are several computer simulations of particulate recording media, based on models of their microscopic behaviour. The simulations were used to probe the macroscopic behaviour of the systems and relate this to the microscopic models used. The four areas covered are: Time dependence in magnetic systems. A refinement of an established model of a perpendicular thin film. Results have been obtained for the logarithmic time dependence of the magnetisation together with the temperature dependence of both the remanent magnetisation and the viscosity coefficient. These all showed good agreement with previous work as did the results for the magnetisation decay of a small area of film. The linear transverse susceptibility of particulate recording media. Calculations have been made for a set of non-interacting Stoner-Wohlfarth particles. The introduction of both angular and magnitude distributions for the anisotropy yielded some very interesting results. Calculations were made for a range of differently textured systems. The results compare very favourably with some measurements obtained by industrial co-workers and explain some of the earlier results obtained by other workers. It is postulated that the transverse suscceptibility measurement may hold information about the magnetisation reversal mechanisms in these media. The effects of texture in the case of the FMR response. An established model was refined to include the effects of texture. Measurements were more closely simulated by adjusting of the texture to be more realistic. Results are shown comparing the measured response with the calculations. A related technique to the linear transverse susceptibility is developed in terms of a non-linear transverse susceptibility. Results of calculations suggest that this may be a useful singular point detection technique for obtaining anisotropy values. Further, since this technique has a high angular selectivity, it would probably be a useful tool in probing the texture of mat en a! s. A comparison between the linea.r transverse susceptibility and FMR responses is presented along with an overall discussion of the results obtained.

Several novel mixed ionic-electronic conductors (MIECs) based on the oxygen interstitial diffusion mechanism have been developed and studied for the past decade. This kind of materials has the potential to be used as the cathode of intermediate temperature solid oxide fuel cell (IT-SOFC) systems. Among them, the La2NiO4+δ Ruddlesden-Popper (RP) and the CeNbO4+δ fergusonite materials attract particular attention due to their interesting properties. These materials are capable to take extra oxygen and become oxygen ion conductor without any aliovalent doping. To understand more about their catalytic and diffusion behaviours, atomistic computer simulations using both classical pair potentials and density functional theory (DFT), as well as classical thermodynamics calculations, have been carried out on these materials. The La2NiO4+δ material has recently been discovered to exhibit a La-only surface. This observation disagrees with previous theoretical study which suggest a Ni-terminated surface and challenges the traditional belief that the oxygen reduction takes place on La2NiO4+δ's exposed Ni sites. Our hybrid DFT surface energy calculation suggests that the most stable (001) La2NiO4 surface is La terminated. Previous theoretical surface study's results may not appropriately reflect the room temperature La2NiO4 sample. To explain the La-only surface observed on the polycrystalline La2NiO4+δ samples, a thermodynamic decomposition based hypothesis is proposed. The evaluation of La2NiO4's stability with respect to the higher ordered RP materials (La3Ni2O7, La4Ni3O10 and LaNiO3) shows that the decomposition of La2NiO4 is thermodynamically favourable and supports our hypothesis. The defective La-Ni-O RP material's decomposition thermodynamics are estimated with the help of hybrid DFT calculations and agree with our hypothesis too. The decomposition hypothesis predicts a Ni-enriched subsurface layer and agrees with the experiments. The La vacancy formation energy and diffusion barriers are calculated with hybrid DFT methods, as these results may provide hints on why the thermodynamically favourable decomposition only limits to the surface of La2NiO4. Before these calculations, different input La2NiO4 phases and magnetic orderings are evaluated to check their impacts on simulation results' reliability. We have shown that the choice of input structure and magnetic ordering will have significant effect on the simulation results. Appropriate choice of the input setups is therefore very important to the quality of the simulation. Based on the evaluation, La2NiO4's room temperature Fmmm phase and gx type anti-ferromagnetic ordering are selected for the DFT calculations. The La vacancy formation energy and diffusion barrier calculated are compared to the oxygen defect formation energy and related diffusion barrier. Possible relations between the La2NiO4 decomposition reaction and the La vacancy's formation/diffusion energetics are also discussed. Similar to the La2NiO4+δ, the CeNbO4+δ fergusonite material has a range of different oxygen enriched phases and is able to conduct oxygen ions. Its characteristic complex modulated structures also attracts much attentions. The details of its modulated oxygen-rich structures remained unclear for a long time. Recently the CeNbO4.25 modulated structure has been fully solved and opens opportunity for further theoretical studies. Molecular dynamics simulations using classical pair potentials are carried out on the stoichiometric CeNbO4 and the modulated CeNbO4.25 materials to study the oxygen diffusion pathways for the first time. The calculated oxygen ions' mean-squared displacements clearly show that the stoichiometric CeNbO4 exhibits no oxygen self-diffusion up to 1773 K. In addition, the oxygen diffusion coefficients in the CeNbO4.25 are calculated and plotted over a range of temperatures. Three regions with different oxygen diffusion behaviours are found from the graph and the oxygen diffusion activation energies are calculated for each of the regions. The differences and similarities between the computed oxygen diffusion behaviours and the experimentally observed oxygen diffusion behaviours are explored and discussed. Finally, the oxygen diffusion trajectory snapshots are taken from the MD simulation. The exact oxygen diffusion pathways and possible diffusion anisotropicities of the CeNbO4.25 material are analysed and discussed based upon the diffusion trajectories.

The magnetic properties of nanometre-scale structures are of fundamental scientific interest and have the potential to play a major role in future data storage technologies. In particular, arrays of small magnetic elements, also called bit-patterned media, are one of the most promising candidates for the future generation of data storage devices. In this thesis we study potential bit patterned element geometries which are below 1 micrometre in size. Their magnetic behaviour is hard to predict using analytical methods and computer simulations are the principal tool for in-depth analysis. The relevant micromagnetic equations are solved using the combined Finite Element/Boundary Element method, and finite differences. Patterned media are (quasi) periodic arrangements of identical objects, with each object typically representing one bit. While one or some of these objects can be simulated with today’s simulation capabilities, the investigation of arrays with hundreds of objects requires novel simulation methods. To deal with such large arrays we introduce and evaluate the new “macro geometry” approach. In most real samples this is superior to using conventional periodic boundary conditions as it takes account of the macroscopic shape of the sample. The micromagnetic simulation package Nmag developed at Southampton has been extended to provide the macro geometry capabilities, and subsequently used to study demagnetising effects between the elements of triangular ring arrays. We find that in a square array of 50-nm size triangular elements these effects are governed by the first and second nearest neighbours and can be considered negligible when the spacing between the rings is larger than 30 nm. We also study the transport properties via the Anisotropic Magneto Resistance (AMR) signal of connected rings arrays using the multi-physics features of Nmag. The simulations use a self-consistent approach to determine the AMR values, a technique able to explain experimental AMR measurements of real structures. We also show how the spatially varying current distribution affects the computation of the AMR values and found that the uniform current model, sometimes used in the study of AMR effects, is a very inaccurate approximation and can easily lead to qualitatively wrong results.

This paper describes the use of corporate decision and strategy simulations as a decision-support instrument under varying market conditions in the tourism industry. It goes on to illustrate this use of simulations with an experiment which investigates how successful different market segmentation approaches are in destination management. The experiment assumes a competitive environment and various cycle-length conditions with regard to budget and strategic planning. Computer simulations prove to be a useful management tool, allowing customized experiments which provide insight into the functioning of the market and therefore represent an interesting tool for managerial decision support. The main drawback is the initial setup of a customized computer simulation, which is time-consuming and involves defining parameters with great care in order to represent the actual market environment and to avoid excessive complexity in testing cause-effect-relationships. (author's abstract)
Series: Report Series SFB "Adaptive Information Systems and Modelling in Economics and Management Science"

In this work we have studied energy flow in acoustic billiards, focusing on irregular billiards with and without current effects. The open systems were modeled with an imaginary potential as a source and drain. We have used the finite difference method to model the billiards. General features of the systems are reported and effects of the measuring probe on the wave function are discussed.

A novel and efficient methodology is developed for computer simulations of realistic two-dimensional (2D) and three-dimensional (3D) microstructures. The simulations incorporate realistic 2D and 3D complex morphologies/shapes, spatial patterns, anisotropy, volume fractions, and size distributions of the microstructural features statistically similar to those in the corresponding real microstructures. The methodology permits simulations of sufficiently large 2D as well as 3D microstructural windows that incorporate short-range (on the order of particle/feature size) as well as long-range (hundred times the particle/feature size) microstructural heterogeneities and spatial patterns at high resolution. The utility of the technique has been successfully demonstrated through its application to the 2D microstructures of the constituent particles in wrought Al-alloys, the 3D microstructure of discontinuously reinforced Al-alloy (DRA) composites containing SiC particles that have complex 3D shapes/morphologies and spatial clustering, and 3D microstructure of boron modified Ti-6Al-4V composites containing fine TiB whiskers and coarse primary TiB particles. The simulation parameters are correlated with the materials processing parameters (such as composition, particle size ratio, extrusion ratio, extrusion temperature, etc.), which enables the simulations of rational virtual 3D microstructures for the parametric studies on microstructure-properties relationships. The simulated microstructures have been implemented in the 3D finite-elements (FE)-based framework for simulations of micro-mechanical response and stress-strain curves. Finally, a new unbiased and assumption free dual-scale virtual cycloids probe for estimating surface area of 3D objects constructed by 2D serial section images is also presented.

Designing porous materials capable of hosting molecular species in cavities of molecular dimensions is an important goal in materials science and engineering. Contrary to the small, transient cavities that exist in the interstitial voids between the molecules of any conventional liquid, a "porous liquid" possesses well-defined, permanent voids inherent to the molecular architecture of the liquid molecules. The efficacy of such a microporous liquid lies in the ability of these intramolecular cavities to selectively house molecular guests. In this thesis we use computer simulations to embark on the first investigation into the microporous nature of potential "porous liquids", and establish whether this novel concept is viable. Simulations studies, using models as an approximation to the geometry of cage-like molecules, suggest that it is possible to design a system that exhibits permanent microporosity in the liquid phase. By calculating the distribution of cavity sizes in the liquid samples and simulating gas absorption experiments, we also establish that an existing system (synthesised by our experimental collaborators) is a potential candidate as the first microporous liquid.

Computer simulations for investigating protein folding and evolution are presented. In chapter 1, an all-atom model with a knowledge-based potential is used to study the folding kinetics of Formin-Binding protein. We study the folding kinetics by performing Monte Carlo simulations. We examine the order of formation of two beta-hairpins, the folding mechanism of each individual beta-hairpin, and transition state ensemble (TSE) and compare our results with experimental data and previous computational studies. Further, a rigorous Pfold analysis is used to obtain representative samples of the TSEs showing good quantitative agreement between experimental and simulated phi values.

The martensitic transition in zirconium is a first order solid-to-solid transition which transforms from body-centred-cubic (bcc) to hexagonal close-packed (hcp) structure. By using a Finnis-Sinclair type many-body potential derived for zirconium and molecular dynamics (MD) methods, a large number of simulations have been studied with the implementation of Nosé-Hoover thermostat and Parrinello-Raman scheme. We found that the transition is a result of the instability of a transverse N-point phonon in the bcc lattice which can be stabilised by the large fluctuation of the anharmonic effect above the transition temperature via extra vibrational entropy. The transition temperature in our calculations is 1,333K. With the concept of 'local atomic structure', the microstructure can be studied at the atomic level. The kinetics of the transition is dominated by the strain energy. Stripes of (10?1) twins are formed as a consequence. The twins contain some stacking faults. Stacking faults play a major role in the splitting of partial dislocations in twins. After the martensitic microstructure is deformed, we found that the twinning deformation occurs by the aid of partial dislocations and the interchange of neighbouring atoms. The interchange of neighbouring atoms leads to the term 'plastic atoms'. We believe that these plastic atoms cause a microscopic irreversible process and the absent of the shape memory effect in zirconium.

Includes bibliographical references (leaves 187-210).
The structure and dynamics of dendrimers in solution are studied through nanosecond atomistic Molecular Dynamics (MD) simulations of explicitly solvated Fréchet dendrimers, generations G1 to G5. The properties of these dendrimers are investigated in four solvent invironments: vacuum and water (representatives of poor solvents), and tetrahydrofuran (THF) and chloroform (representatives of good solvents). To establish the quality of the solvent on the conformation of the dendrimer, additional nanosecond MD simulations fo the dendrimers are performed, from both inititally folded and unfolded conformations.

Les membranes fluorées sont utilisées en particulier dans les dénommées piles à combustible à membrane électrolyte polymère. Grâce à sa grande mobilité en protons, le célèbre ionomer Nafion® (Dupont) est un matériau de référence pour les applications liées aux piles à combustible. En présence d’eau ou d’autres solvants hydrophiles la membrane se sépare en une matrice polymérique hydrophobe et une sous-phase aqueuse contenant des clusters d’eau et ions, dont les tailles et la connectivité augmente quand la quantité d’eau augmente [1]. Quelle est la morphologie du Nafion et la structure du solvant, dans de tels systèmes?Il a été récemment montré [2] sur des simulations de large systèmes que plusieurs modèles morphologiques reproduisent les données expérimentales de diffusion, évoquant l’incapacité des mesures de diffusion seules à élucider la véritable structure du Nafion.Néanmoins, un modèle ’aléatoire’ décrit dans [2], c’est à dire l’unique modèle étudié sans présumer d’une structure initiale particulière, n’a pas pu reproduire les données expérimentales.Générer en simulations moléculaires des configurations du système qui soient vraiment décorrélées de la configuration initiale reste un vrai défi statistique. Les échelles de temps réalisables ne permettent simplement pas d’obtenir des mouvements significatifs du polymère (comme des transitions de conformations, repliements de chaînes, etc.). Nous proposons ainsi dans cette étude un nouveau modèle de Nafion à morphologie aléatoire. Un algorithme récemment développé est utilisée pour générer des chaînes de Nafion avec des chemins et des points de départ aléatoires. Une différence majeure avec le modèle aléatoire dans [2] est que nous ne construisons pas nos systèmes à une densité proche de la densité finale. Pour ne pas démarrer avec des chaînes trop enchevêtrées, les systèmes sont initialement préparés à une densité en dessous de la référence expérimentale. La densité après équilibration est de nouveau proche de l’expérience. Bien qu’il soit facilement envisageable d’améliorer les nouveaux algorithmes, nous démontrons ici qu’avec la présente version plusieurs séries de configurations compatibles avec les données expérimentales de diffusion disponibles peuvent être générées et équilibrées. Douze large systèmes de Nafion à morphologie aléatoire sont construits avec des positions initiales des atomes ainsi que des quantités d’eau et des longueurs de chaînes (Nafion/Hyflon) différentes. Ils sont équilibrés puis simulés sur plusieurs dizaines de nanosecondes. Après équilibration, les structures sont, comme indiqué ci-dessus,compatibles avec les données expérimentales de diffusion. En plus nous étudions un modèle ressemblant à celui de Schmidt-Rohr and Chen [3], c’est à-dire le plus récent modèle morphologique. Avec ce modèle, les données expérimentales sont également reproduites de manière satisfaisante, d’où la prolongation du débat sur la structure du Nafion. La cohésion entre les valeurs calculées et celles mesurées expérimentalement incite à des analyses plus en détails de ces configurations obtenues. Nous caractérisons et analysons les structures locales, intermédiaires et à grande échelle avec divers paramètres structuraux et distributions des tailles de domaines. Nous calculons donc, par exemple, des fonctions de distribution radiale (rdf), des facteurs de structure (S(q)) totaux et partiels tout comme des nombres et des tailles de clusters hydrophiles (selon la définition d’un cluster). La dynamique de diverses espèces dans le système est également examinée,par exemple au travers des déplacements carrés moyens (msd) et des coefficients de diffusion. Ces simulations sont probablement à la limite de ce qui est réalisable aujourd’hui avec des simulations ’full-atom’ du type MD. Nous espérons que ce travail fera avancer le débat sur la structure et la dynamique de ces matériaux importants
Perfluorinated membranes are used in particular in polymer electrolyte fuel cells(PEFC). The well-known ionomer Nafion® (Dupont) is, due to its high proton mobility,a reference material for fuel cell applications. In water or other hydrophilic solvents themembrane segregates into a hydrophobic backbone matrix and a hydrophilic sub-phasecontaining clusters of both water and ions, where the cluster sizes and connectivity increasewith increasing water content [1].What is the Nafion morphology and the structure of the solvent in such systems? It hasbeen shown recently [2] on large simulated systems that several morphological modelsfit the experimental scattering data, suggesting the inability of scattering experimentsalone to elucidate the true structure of Nafion. However, a ’random’ model describedin [2], i.e. the only explored model that did not assume a particular initial structure,could not reproduce the experimental data.It remains a real computational challenge to generate in molecular simulations systemconfigurations which are really decorrelated from the initial one. The time scales thatcan be achieved simply do not allow to obtain significant motions of the polymer (e.g.conformational changes, folding, etc.). We thus propose in this work a new randommodel of Nafion. A newly developped algorithm is used to generate Nafion chains withrandom growth paths and random starting points. A significant difference with therandom model in [2] is that we do not build our systems at a density close to the finalone. In order not to start with too much entangled chains, the systems are initiallybuilt at a density below the experimental one. The density after equilibration is againclose to the experimental one.Even though further improvements of the new algorithms can easily be envisaged,we demonstrate here that with the present version several sets of configurations thatare compatible with the available scattering data can be generated and equilibrated.Twelve large random Nafion systems are built with different initial positions of theatoms as well as different water contents and side chain lengths (Nafion/Hyflon). Theyare equilibrated and then simulated for several ten nanoseconds. After equilibration,the structures are, as mentioned, compatible with the experimental scattering data. Inaddition we study a model similar to the one by Schmidt-Rohr and Chen [3], i.e. thenewest morphological model of Nafion. The experimental scattering data are also satisfactorilyreproduced with this model, hence, the prolonged debate over the structureof Nafion.This agreement gives confidence that a more detailed analysis of the so-obtained configurationsis scientifically warranted. We characterize and analyze the local, intermediateand large-scale structures by various structural parameters and domain size distributions.We therefore compute, for example, radial distribution functions (rdf), total andpartial structure factors (S(q)) as well as numbers and sizes of hydrophilic clusters (dependingon the definition of a cluster). The dynamics of various species in the systemis also investigated, e.g. via the computation of the mean square displacements (msd)and the self-diffusion coefficients. These simulations are probably at the limit of whatcan today be achieved with all-atom molecular simulations of the MD type. We hopethat this work will advance the ongoing debate on the structure and dynamics of theseimportant materials
Perfluorierte Membranen werden insbesondere in Polymerelectrolyt-Brennstoffzellen(PEFC) eingesetzt. Das wohlbekannte Ionomer Nafion® (Dupont) ist wegen seinerhohen Protonenbeweglichkeit ein Referenzmaterial für solche Anwendungen in Brennstoffzellen.Die Membran separiert in Wasser oder anderen hydrophilen Lösungsmittelin eine hydrophobe Polymermatrix und eine hydrophile Subphase, die Cluster mitWasser und Ionen enthält. Dabei vergroeßern sich die Ausdehnung der Cluster und ihreKonnektivität mit zunehmendem Wassergehalt [1].Welche ist die Morphologie des Nafions und die Struktur des Lösungsmittels in diesenSystemen? Es ist jüngst anhand großer simulierter Systeme gezeigt worden [2], dassmehrere morphologische Modelle die experimentellen Streudaten wiedergeben können,was nahelegt, dass solche Streudaten alleine nicht geeignet sind, die wahre Strukturdes Nafion aufzudecken. Ein in [2] beschriebenes ’Zufallsmodell’, d.h. das einzigeder untersuchten Modelle, das keine besondere Anfangsstruktur annahm, konnte dieexperimentellen Daten allerdings nicht wiedergeben.In molekularen Computersimulationen Konfigurationen zu erzeugen, die wirklich nichtmehr mit der angenommenen Anfangskonfiguration korreliert sind, bleibt eine echteHerausforderung. Die erreichbaren Zeitskalen sind zu kurz, um eine signifikante Bewegungdes Polymers (z.B Konformationsänderungen, Faltungen, usw.) zuzulassen. Indieser Arbeit wird daher ein neues Zufallsmodell für Nafion vorgestellt. Ein neuentwickelterAlgorithmus erzeugt Nafionketten mit zufälligem Wachstumspfad ausgehendvon zufälligen Anfangspunkten. Ein signifikanter Unterschied zu dem Zufallsmodellvon [2] ist, dass hier nicht versucht wird, die Systeme bei einer Dichte vergleichbarder experimentellen Dichte aufzubauen. Anstattdessen werden die Systeme, um alzustarkes Verknäuelung zu vermeiden, anfangs bei einer deutlich kleineren Dichte erzeugt.Nach äquilibrierung ist die Systemdichte wieder in etwa gleich der experimentellen.Wiewohl weitere Verbesserungen des neu Algorithmuses leicht ins Auge gefaßt werdenkönnen, so kann hier doch gezeigt werden, dass mit der gegenwärtigen VersionKonfigurationen erzeugt und äquilibriert werden können, die mit den verfügbarenStreudaten kompatibel sind. Zwölf große Nafion Zufallssysteme, mit verschiedenenAnfangspositionen der Atome, verschiedenem Wassergehalt und Längen der Seitenketten(Nafion/Hyflon) werden aufgebaut. Diese werden äquilibriert und mehrerezehn Nanosekunden lang simuliert. Nach der äquilibrierung sind die Strukturen, wieerwähnt, kompatibel mit den experimentellen Streudaten. Weiterhin wird ein Modellähnlich dem von Schmidt-Rohr und Chen [3], d.h. dem neuesten morphologischen Modellfür Nafion, studiert. Auch hier werden die experimentellen Streudaten zufriedenstellendwiedergegeben, daher die weiterhin bestehende Debatte über die Struktur desNafion.Die gefundenen übereinstimmungen lassen darauf vertrauen, dass eine detaillierte Analyseder simulierten Konfigurationen wissenschaftlich sinnvoll ist. So wird die Strukturder Systeme auf verschiedenen Längenskalen charakterisiert, zum Beispiel durch radialePaarverteilungsfunktionen (rdf), totale und partielle Strukturfaktoren (S(q)) sowieAnzahl- und Größenverteilungen hydrophiler Cluster (abhängig von der Definition einesClusters). Die Dynamik einzelner Spezies im System wird ebenfalls untersucht, zumBeispiel durch die Berechnung der mittleren quadratischen Verschiebungen (msd) undder Selbstdiffusionskoeffizienten. Diese Simulationen sind wahrscheinlich an der Grenzedessen, was heute mit ’all-atom’ molekularen MD-Simulationen möglich ist. Ich vertrauedarauf, dass diese Arbeit dennoch einen Fortschritt in der aktuellen Debatte überdie Struktur und Dynamik dieser wichtigen Materiale darstellt

Les interactions protéine-protéine (IPPs) médient la signalisation cellulaire. Leur ingénierie peut fournir des informations et conduire au développement de molécules thérapeutiques. Les domaines PDZ sont des médiateurs importants de IPPs. Elles lient les 4--10 résidus C-terminaux de protéines cibles. Elles lient aussi les peptides correspondants, qui peuvent servir de systèmes modèles ou d'inhibiteurs. Nous avons développé deux approches computationnelles et les avons appliquées au domaine PDZ de la protéine Tiam1, un facteur d'échange pour la protéine Rac, impliqué dans la protrusion neuronale. Sa cible est la protéine Syndecan1. Des affinités expérimentales sont connues pour le peptide C-terminal, noté Sdc1, et plusieurs mutants; elles ont servi pour tester les calculs. Nous avons d'abord développé une méthode de dessin computationnel haut débit. Une simulation Monte Carlo est faite où les chaines latérales de la protéine et du peptide peuvent changer de conformères et certaines positions peuvent muter. Le solvant est implicite. Le paysage énergétique est aplati par la méthode adaptative de Wang-Landau, de sorte qu'un vaste ensemble de variantes est échantillonné. Effectuant des simulations distinctes du complexe et du peptide seul nous avons obtenu les énergies libres relatives d'association de 75,000 variantes en heure CPU sur une machine de bureau. Les valeurs sont compatibles avec les quelques données expérimentales disponibles. Ensuite, nous avons développé une approche beaucoup plus détaillée et réaliste. Soluté et solvant sont décrits par un champ de force atomique, qui représente explicitement la polarisation électronique: le champ de force Drude de Charmm. La polarisabilité peut être importante car les résidus de l'interface PDZ:peptide passent, lors de l'association, d'un environnement riche en solvant à un autre pauvre en solvant. Nous avons fait des simulations alchimiques d'énergie libre pour comparer quatre variantes du peptide qui diffèrent par une ou deux chaines latérales ioniques. Les résultats sont en bon accord avec l'expérience. Les champs de force additifs Charmm et Amber, qui représentent la polarisabilité implicitement, donnent un moins bon accord. Ces calculs sont le premier exemple de simulations alchimiques d'énergies libre d'association relatives protéine: ligand avec un champ de force polarisable. Enfin, pour une modélisation future de peptides phosphorylés, nous avons étendu le champ de force Drude pour inclure le méthyl phosphate et la phospho tyrosine. Il en résulte un excellent accord avec les affinités expérimentales phosphate: magnésium
Protein-protein interactions (PPIs) regulate complex signaling networks in eukaryotic cells. Many binding events between several protein domains transfer information through communication pathways. Disrupting or altering the equilibrium between PPIs plays an important role inseveral diseases and the inibition of targeted PPIs is a recognized strategy for computational drug design. In the present thesis we focused on PDZ domains, which are among the most widespread signaling domains. PDZs recognize the 4-10 C-terminal amino acids of their target proteins as well as the corresponding peptides in isolation. We studied PDZ:peptide binding for the Tiam1 protein, which is a Rac GTP exchange factor involved in neuronal protrusion and axon guidance. Tiam1 activity modulates signaling for cell proliferation and migration, whose dysregulation increases growth of metastatic cancers. Its natural binder peptide is Syndecan1 (Sdc1), composed of 8 amino acids. Its last 5 Cter residues drive interactions in the binding pocket. Experimental affinities for several mutants of Sdc1 and in the protein domain constitute a complete dataset to study many ionic interactions with molecular simulations. These calculations are still challenging, despite the dramatic improvement of biomolecular modelling in the 1990's and 2000's. Upon binding, residues are transferred from a solvent-exposed environment to a solvent-poor one. This is expected to change the electron distribution within residues and nearby solvent molecules. Comparing ligands that differ by one or more ionic side-chain mutations, more sophisticated force fields where electronic polarizability is treated explicitly may be required. We developed and tested both Computational Protein Design (CPD) models and more precise free energy calculation methods based on polarizable molecular dynamics. We developed a general, high-througtput CPD protocol to optimize protein:peptide binding. The model has been implemented in on our in-house CPD package Proteus ( Simonson et al, 2014) and has been tested computing relative binding affinities for many variants of the Tiam1:Sdc1 complex. Monte Carlo sampling of equilibrium distributions of protein sequences is performed using an adaptive bias potential which flattens the energy landscape in sequence space and allows to estimate binding affinities for thousands of protein variants in limited CPU time (~1hour). We also improved our CPD implicit solvent model, implementing a more realistic description of the solute-solvent dielectric boundary. The new method, called Fluctuating Dielectric Boundary (FDB) showed a systematic improvement in the prediction of acid:base constants of several proteins. Promising results were also obtained for the complete sequence redesign of three PDZ domains. In the second part of this work we studied Tiam1:peptide affinities with more sophisticated models, based on free energy simulations with the Drude Polarizable Force field (DrudeFF). We first computed relative binding free energies for charge mutations in the Tiam1:Sdc1 complex, obtaining a clear improvement respect to equivalent calculations performed using two additive force fields. We applied the well-enstablished Dual Topology Approach: to our knowledge, this was the first example of such a calculation for a protein:peptide complex with uses the DrudeFF. Then we went on, developing the Drude polarizable models for methyl phosphate (MP) and phospho tyrosine (pTyr). We were interested in the change in binding affinity associated with phosphorylation of a Tyrosine residue of Sdc1, but Drude pTyr parameters were not yet developed. We tested our new phosphate parameters studying standard binding free energies between MP and magnesium (Mg2+) in water solution. Results showed a good agreement with experiment, improving previous calculations performed using additive force field