Biomolecular Modeling is playing an important role in many practical applications such as biotechnology and structure-based drug design. One of the essential requirements of Biomolecular modeling is an accurate description of the solvent (water). The challenge is to make this description computationally facile that is reasonably fast, simple, robust and easy to incorporate into existing software packages. The most rigorous procedure to model the effect of aqueous solvent is to explicitly model every water molecule in the system. For many practical applications, this approach is computationally too intense, as the number of required water atoms is on average one order of magnitude larger than the number of atoms of the molecule of interest. Implicit solvent models, in which solvent molecules are represented by a continuum function, have become a popular alternative to explicit solvent methods as they are computationally more efficient. The Generalized Born (GB) implicit solvent has become quite popular due to its relative simplicity and computational efficiency. However, recent studies showed serious deficiencies of many GB variants when applied to Biomolecular Modeling such as an over- stabilization of alpha helical secondary structures and salt bridges. In this dissertation we present two new GB models aimed at computing solvation properties with a reasonable compromise between accuracy and speed. The first GB model, called NSR6, is based on a numerically surface integration over the standard molecular surface. When applied to a set of small drug-like molecules, NSR6 produced an accuracy, with respect to experiments, that is essentially at the same level as that of the expensive explicit solvent treatment. Furthermore, we developed an analytic GB model, called AR6, based on an approximation of the volume integral over the standard molecular volume. The accuracy of the AR6 model is tested relative to the numerically exact NSR6. Overall AR6 produces a good accuracy and is suitable for Molecular Dynamics simulations which is the main intended application. / Ph. D.

<p>Accurate calculations of free energies for molecular association and solvation are important for the understanding of biochemical processes, and are useful in many pharmaceutical applications. In this thesis, molecular dynamics (MD) simulations are used to calculate thermodynamic properties for solvation and ligand binding.</p><p>The thermodynamic integration technique is used to calculate p<i>K</i>_a values for three aspartic acid residues in two different proteins. MD simulations are carried out in explicit and Generalized-Born continuum solvent. The calculated p<i>K</i>_a values are in qualitative agreement with experiment in both cases. A combination of MD simulations and a continuum electrostatics method is applied to examine p<i>K</i>_a shifts in wild-type and mutant epoxide hydrolase. The calculated p<i>K</i>_a values support a model that can explain some of the pH dependent properties of this enzyme.</p><p> Development of the linear interaction energy (LIE) method for calculating solvation and binding free energies is presented. A new model for estimating the electrostatic term in the LIE method is derived and is shown to reproduce experimental free energies of hydration. An LIE method based on a continuum solvent representation is also developed and it is shown to reproduce binding free energies for inhibitors of a malaria enzyme. The possibility of using a combination of docking, MD and the LIE method to predict binding affinities for large datasets of ligands is also investigated. Good agreement with experiment is found for a set of non-nucleoside inhibitors of HIV-1 reverse transcriptase.</p><p>Approaches for decomposing solvation and binding free energies into enthalpic and entropic components are also examined. Methods for calculating the translational and rotational binding entropies for a ligand are presented. The possibility to calculate ion hydration free energies and entropies for alkali metal ions by using rigorous free energy techniques is also investigated and the results agree well with experimental data.</p>

O método de equalização da eletronegatividade (Electronegativity Equalization Method, EEM), fundamentado em teoria do funcional da densidade eletrônica, foi combinado à aproximação de Born generalizada para moléculas (Generalized Born, GB), e denominado GBEEM (Dias et al., 2002). Os momentos de dipolo permanente no vácuo e em meio condensado (constante dielétrica ~ 80), e distribuições de cargas atômicas, mostraram boa concordância com modelo SM5.4 baseado em cargas CM1 em nível PM3 (12 moléculas, correspondendo a 29 cargas atômicas). Este resultado é interessante devido à simplicidade inerente do GBEEM e seu baixo custo computacional. Uma nova parametrização das durezas e eletronegatividades foi feita com o objetivo de melhorar a distribuição de cargas atômicas em moléculas isoladas em relação ao modelo CM1. Um conjunto de 250 estruturas/cargas PM3/CM1 de moléculas neutras pertencentes a 13 funções orgânicas foi utilizado como alvo na parametrização, utilizando uma metodologia Algoritmo Genético/Simplex de pesquisa de mínimos (Menegon et al., 2002). Boa concordância entre os modelos foi obtida. A validação da parametrização e do EEM foi efetuada usando moléculas bifuncionais (tetrapeptídeo e trisacarídeo) mostrando também boa concordância e robustez. Entretanto, a análise do momento de dipolo permanente das 250 moléculas mostrou uma séria limitação do EEM, e portanto do GBEEM, apesar da boa concordância entre as cargas EEM e CM1. O EEM superestimou os momentos de dipolo. Tal fato pode decorrer de vários fatores, dentre os quais, o truncamento da expansão nas cargas atômicas e ausência de tratamento explícito de interação de troca (exchange). Foi sugerida uma aproximação que restringe a transferência de carga entre grupos na molécula que contornou a limitação do método na predição de momentos de dipolo no vácuo e meio condensado (Shimizu et al., 2004). Com base nos recentes resultados, foi desenvolvido um modelo de solvatação baseado no GBEEM e no modelo de Floris-Tomasi. A calibração foi feita com um conjunto de 62 moléculas neutras (13 grupos funcionais) tendo como alvo as energias livres de hidratação experimentais. Os resultados apresentaram um desvio médio absoluto de 0,71 kcal/mol em relação aos valores experimentais. / The electronegativity equalization method (EEM), founded on density functional theory (DFT), has been combined to the generalized Born approximation (GB) for molecules, and called GBEEM (Dias et al., 2002). The permanent dipole moment in vacuum and condensed phase (dieletric constant ~ 80), and atomic charges distributions, have shown good agreement with SM5.4 solvation model based on CM1 charges at PM3 level (12 molecules, corresponding to 29 atomic charges). This result is interesting due the simplicity of GBEEM and its low computational cost. A new parameterization of the hardness and electronegativities was done with the aim to improve the atomic charges distribution on isolated molecules in comparison to CM1 model. The training set with 250 PM3/CM1 structures/charges of neutral molecules in 13 different organic functions was employed as target in the parameterization. A new optimization approach composed of Genetic and Simplex algorithms was used to fit parameters (Menegon et al., 2002). Good agreement between the models was found. The validation of parameterization and EEM was done using bifunctional molecules (tri-glucose and tetra-peptide) showing good agreement and robustness. However, analysis of permanent dipole moments of 250 molecules shown a serious caveat of EEM and GBEEM, beside the good agreement between EEM and CM1 charges. EEM has overestimated the dipole moments. Such result may be due to the truncated expansion in atomic charges and lacking of explicit treatment of exchange interaction. A new approximation was proposed constraining the charge transfer between groups within the molecule. This approximation corrected the caveat of EEM in the prediction of dipole moments in vacuum and condensed phase (Shimizu et al., 2004). Based on these results, a new solvation model was developed founded in GBEEM and Floris-Tomasi model. The parameterization was done with a training set of 62 neutral molecules (13 functional groups) and experimental hydration free energies as target. This new solvation model has produced a mean absolute deviation, MAD, of 0.71 kcal/mol comparing to experimental data.

Accurate calculations of free energies for molecular association and solvation are important for the understanding of biochemical processes, and are useful in many pharmaceutical applications. In this thesis, molecular dynamics (MD) simulations are used to calculate thermodynamic properties for solvation and ligand binding. The thermodynamic integration technique is used to calculate pKa values for three aspartic acid residues in two different proteins. MD simulations are carried out in explicit and Generalized-Born continuum solvent. The calculated pKa values are in qualitative agreement with experiment in both cases. A combination of MD simulations and a continuum electrostatics method is applied to examine pKa shifts in wild-type and mutant epoxide hydrolase. The calculated pKa values support a model that can explain some of the pH dependent properties of this enzyme. Development of the linear interaction energy (LIE) method for calculating solvation and binding free energies is presented. A new model for estimating the electrostatic term in the LIE method is derived and is shown to reproduce experimental free energies of hydration. An LIE method based on a continuum solvent representation is also developed and it is shown to reproduce binding free energies for inhibitors of a malaria enzyme. The possibility of using a combination of docking, MD and the LIE method to predict binding affinities for large datasets of ligands is also investigated. Good agreement with experiment is found for a set of non-nucleoside inhibitors of HIV-1 reverse transcriptase. Approaches for decomposing solvation and binding free energies into enthalpic and entropic components are also examined. Methods for calculating the translational and rotational binding entropies for a ligand are presented. The possibility to calculate ion hydration free energies and entropies for alkali metal ions by using rigorous free energy techniques is also investigated and the results agree well with experimental data.

O método de equalização da eletronegatividade (Electronegativity Equalization Method, EEM), fundamentado em teoria do funcional da densidade eletrônica, foi combinado à aproximação de Born generalizada para moléculas (Generalized Born, GB), e denominado GBEEM (Dias et al., 2002). Os momentos de dipolo permanente no vácuo e em meio condensado (constante dielétrica ~ 80), e distribuições de cargas atômicas, mostraram boa concordância com modelo SM5.4 baseado em cargas CM1 em nível PM3 (12 moléculas, correspondendo a 29 cargas atômicas). Este resultado é interessante devido à simplicidade inerente do GBEEM e seu baixo custo computacional. Uma nova parametrização das durezas e eletronegatividades foi feita com o objetivo de melhorar a distribuição de cargas atômicas em moléculas isoladas em relação ao modelo CM1. Um conjunto de 250 estruturas/cargas PM3/CM1 de moléculas neutras pertencentes a 13 funções orgânicas foi utilizado como alvo na parametrização, utilizando uma metodologia Algoritmo Genético/Simplex de pesquisa de mínimos (Menegon et al., 2002). Boa concordância entre os modelos foi obtida. A validação da parametrização e do EEM foi efetuada usando moléculas bifuncionais (tetrapeptídeo e trisacarídeo) mostrando também boa concordância e robustez. Entretanto, a análise do momento de dipolo permanente das 250 moléculas mostrou uma séria limitação do EEM, e portanto do GBEEM, apesar da boa concordância entre as cargas EEM e CM1. O EEM superestimou os momentos de dipolo. Tal fato pode decorrer de vários fatores, dentre os quais, o truncamento da expansão nas cargas atômicas e ausência de tratamento explícito de interação de troca (exchange). Foi sugerida uma aproximação que restringe a transferência de carga entre grupos na molécula que contornou a limitação do método na predição de momentos de dipolo no vácuo e meio condensado (Shimizu et al., 2004). Com base nos recentes resultados, foi desenvolvido um modelo de solvatação baseado no GBEEM e no modelo de Floris-Tomasi. A calibração foi feita com um conjunto de 62 moléculas neutras (13 grupos funcionais) tendo como alvo as energias livres de hidratação experimentais. Os resultados apresentaram um desvio médio absoluto de 0,71 kcal/mol em relação aos valores experimentais. / The electronegativity equalization method (EEM), founded on density functional theory (DFT), has been combined to the generalized Born approximation (GB) for molecules, and called GBEEM (Dias et al., 2002). The permanent dipole moment in vacuum and condensed phase (dieletric constant ~ 80), and atomic charges distributions, have shown good agreement with SM5.4 solvation model based on CM1 charges at PM3 level (12 molecules, corresponding to 29 atomic charges). This result is interesting due the simplicity of GBEEM and its low computational cost. A new parameterization of the hardness and electronegativities was done with the aim to improve the atomic charges distribution on isolated molecules in comparison to CM1 model. The training set with 250 PM3/CM1 structures/charges of neutral molecules in 13 different organic functions was employed as target in the parameterization. A new optimization approach composed of Genetic and Simplex algorithms was used to fit parameters (Menegon et al., 2002). Good agreement between the models was found. The validation of parameterization and EEM was done using bifunctional molecules (tri-glucose and tetra-peptide) showing good agreement and robustness. However, analysis of permanent dipole moments of 250 molecules shown a serious caveat of EEM and GBEEM, beside the good agreement between EEM and CM1 charges. EEM has overestimated the dipole moments. Such result may be due to the truncated expansion in atomic charges and lacking of explicit treatment of exchange interaction. A new approximation was proposed constraining the charge transfer between groups within the molecule. This approximation corrected the caveat of EEM in the prediction of dipole moments in vacuum and condensed phase (Shimizu et al., 2004). Based on these results, a new solvation model was developed founded in GBEEM and Floris-Tomasi model. The parameterization was done with a training set of 62 neutral molecules (13 functional groups) and experimental hydration free energies as target. This new solvation model has produced a mean absolute deviation, MAD, of 0.71 kcal/mol comparing to experimental data.

En sjukdom kan många gånger härledas till en kaskadereaktion mellan proteiner, co-faktorer och substrat. Denna kaskadreaktion blir många gånger målet för att behandla sjukdomen med läkemedel. För att designa nya läkemedelsmoleyler används vanligen datorbaserade verktyg. Denna design av läkemedelsmolekyler drar stor nytta av att målproteinet är känt och då framförallt dess tredimensionella (3D) struktur. Är 3D-strukturen känd kan man utföra så kallad struktur- och datorbaserad molekyldesign, 3D-geometrin (f.f.a. för inbindningsplatsen) blir en vägledning för designen av en ny molekyl. Många faktorer avgör interaktionen mellan en molekyl och bindningsplatsen, till exempel fysikalisk-kemiska egenskaper hos molekylen och bindningsplatsen, flexibiliteten i molekylen och målproteinet, och det omgivande lösningsmedlet. För att strukturbaserad molekyldesign ska fungera väl måste två viktiga steg utföras: i) 3D anpassning av molekyler till bindningsplatsen i ett målprotein (s.k. dockning) och ii) prediktion av molekylers affinitet för bindningsplatsen. Huvudsyftena med arbetet i denna avhandling var som följer: i) skapa modeler för att prediktera affiniteten mellan en molekyl och bindningsplatsen i ett målprotein; ii) förfina molekyl-protein-geometrin som skapas vid 3D-anpassning mellan en molekyl och bindningsplatsen i ett målprotein (s.k. dockning); iii) karaktärisera proteiner och framför allt deras sekundärstruktur; iv) bedöma effekten av olika matematiska beskrivningar av lösningsmedlet för förfining av 3D molekyl-protein-geometrin skapad vid dockning och prediktion av molekylers affinitet för proteiners bindningsfickor. Ett övergripande syfte var att använda kemometriska metoder för modellering och dataanalys på de ovan nämnda punkterna. För att sammanfatta så presenterar denna avhandling metoder och resultat som är användbara för strukturbaserad molekyldesign. De rapporterade resultaten visar att det är möjligt att skapa kemometriska modeler för prediktion av molekylers affinitet för bindningsplatsen i ett protein och att dessa presterade lika bra som andra vanliga metoder. Dessutom kunde kemometriska modeller skapas för att beskriva effekten av hur inställningarna för olika parametrar i dockningsprogram påverkade den 3D molekyl-protein-geometrin som dockingsprogram skapade. Vidare kunde kemometriska modeller andvändas för att öka förståelsen för deskriptorer som beskrev sekundärstrukturen i proteiner. Förfining av molekyl-protein-geometrin skapad genom dockning gav liknande och ickesignifikanta resultat oberoende av vilken matematisk modell för lösningsmedlet som användes, förutom för ett fåtal (sex av 30) fall. Däremot visade det sig att användandet av en förfinad geometri var värdefullt för prediktion av molekylers affinitet för bindningsplatsen i ett protein. Förbättringen av prediktion av affintitet var markant då en Poisson-Boltzmann beskrivning av lösningsmedlet användes; jämfört med prediktionerna gjorda med ett dockningsprogram förbättrades korrelationen mellan beräknad affintiet och uppmätt affinitet med 0,7 (R2). / A disease is often associated with a cascade reaction pathway involving proteins, co-factors and substrates. Hence to treat the disease, elements of this pathway are often targeted using a therapeutic agent, a drug. Designing new drug molecules for use as therapeutic agents involves the application of methods collectively known as computer-aided molecular design, CAMD. When the three dimensional (3D) geometry of a macromolecular target (usually a protein) is known, structure-based CAMD is undertaken and structural information of the target guides the design of new molecules and their interactions with the binding sites in targeted proteins. Many factors influence the interactions between the designed molecules and the binding sites of the target proteins, such as the physico-chemical properties of the molecule and the binding site, the flexibility of the protein and the ligand, and the surrounding solvent. In order for structure-based CAMD to be successful, two important aspects must be considered that take the abovementioned factors into account. These are; i) 3D fitting of molecules to the binding site of the target protein (like fitting pieces of a jigsaw puzzle), and ii) predicting the affinity of molecules to the protein binding site. The main objectives of the work underlying this thesis were: to create models for predicting the affinity between a molecule and a protein binding site; to refine the geometry of the molecule-protein complex derived by or in 3D fitting (also known as docking); to characterize the proteins and their secondary structure; and to evaluate the effects of different generalized-Born (GB) and Poisson-Boltzmann (PB) implicit solvent models on the refinement of the molecule-protein complex geometry created in the docking and the prediction of the molecule-to-protein binding site affinity. A further objective was to apply chemometric methodologies for modeling and data analysis to all of the above. To summarize, this thesis presents methodologies and results applicable to structure-based CAMD. Results show that predictive chemometric models for molecule-to-protein binding site affinity could be created that yield comparable results to similar, commonly used methods. In addition, chemometric models could be created to model the effects of software settings on the molecule-protein complex geometry using software for molecule-to-binding site docking. Furthermore, the use of chemometric models provided a more profound understanding of protein secondary structure descriptors. Refining the geometry of molecule-protein complexes created through molecule-to-binding site docking gave similar results for all investigated implicit solvent models, but the geometry was significantly improved in only a few examined cases (six of 30). However, using the geometry-refined molecule-protein complexes was highly valuable for the prediction of molecule-to-binding site affinity. Indeed, using the PB solvent model it yielded improvements of 0.7 in correlation coefficients (R2) for binding affinity parameters of a set of Factor Xa protein drug molecules, relative to those obtained using the fitting software.