Influenza A is one of the most prevalent and significant viral infections worldwide, resulting in annual epidemics and occasional pandemics. Upon infection, antiviral drugs targeting the neuraminidase protein and M2 protein are the only treatment options available. However, the emergence of antiviral drug resistance is concerning, therefore the aim of this work was to identify inhibitor molecules that may bind to highly conserved regions of selected internal influenza A proteins. Sequences of the non-structural protein 1 (NS1), nuclear export protein (NEP) and polymerase basic protein 2 (PB2) from all hosts and subtypes were aligned and the degree of amino acid conservation was calculated based on Valdar's scoring method. Missing parts of the experimental structures were predicted using the I-TASSER server and ligand binding hot spots were identified with computational solvent mapping. Selected binding sites in conserved regions were subjected to virtual screening against two compound libraries using AutoDock Vina and AutoDock 4. Two out of twelve top hit compounds predicted to target the NS1 protein showed capability of reducing influenza A H1N1 replication in plaque reduction assays at concentrations below 100 μM, although the target protein and mechanism of action could not be confirmed. For the NEP, conservation analysis was based on 3000 sequences and binding hot spots were located in common areas amongst three structures. Docking results revealed predicted binding affinities of up to -8.95 kcal/mol, and conserved amino acid residues interacting with top compounds include Arg42, Asp43, Lys39, Ile80, Gln101, Leu105, and Val109. For the PB2 protein, conservation analysis was based on ~12,000 sequences and fifteen potential binding hot spots were identified. Docking results revealed predicted binding affinities of up to -10.3 kcal/mol, with top molecules interacting with the highly conserved residues Gln138, Gly222, Ile539, Asn540, Gly541, Tyr531 and Thr530. The findings from this research could provide starting points for in vitro experiments, as well as the development of antiviral drugs that function to inhibit influenza A replication without leading to resistance.

<p dir="ltr">Proteins and their interactions with each other, with nucleic acids, and with other molecules are foundational to all known forms of life. The three-dimensional structures of these interactions are an essential component of a comprehensive understanding of how they function. Molecular-biological hypothesis formulation and rational drug design are both often predicated on a particular structure model of the molecule or complex of interest. While experimental methods capable of determining atomic-detail structures of molecules and complexes exist, such as the popular X-ray crystallography and cryo-electron microscopy, these methods require both laborious sample preparation and expensive instruments with limited throughput. Computational methods of predicting complex structures are therefore desirable if they can enable cheap, high-throughput virtual screening of the space of biological hypotheses. Many common biomolecular contexts have largely been blind spots for predictive modeling of complex structures. In this direction, docking methods are proposed to address extreme conformational change, nonuniform environments, and distance-geometric priors. Flex-LZerD deforms a flexible protein using a novel fitting procedure based on iterated normal mode decomposition and was shown to construct accurate complex models even when an initial input subunit structure exhibits extreme conformational differences from its bound state. Mem-LZerD efficiently constrains the docking search space by augmenting the geometric hashing data structure at the core of the LZerD algorithm and enabled membrane protein complexes to be efficiently and accurately modeled. Finally, atomic distance-based approaches developed during modeling competitions and collaborations with wet lab biologists were shown to effectively integrate domain knowledge into complex modeling pipelines.</p>

The goal of the ENCODE project has been to characterize regulatory elements in the human genome, such as regions bound by transcription factors (TFs), regions of open chromatin and regions with altered histone modifications. The ENCODE consortium has performed a large number of whole-genome experiments to measure TF binding, chromatin accessibility, gene expression and histone modifications, on a multitude of cell types and conditions in both human and mouse. In this dissertation I describe the analysis of numerous datasets comprising 66 epigenomes, chromatin accessibility and expression data across twelve tissues and seven time points, during mouse embryonic development. We defined chromatin states using histone modification data and performed integrative analysis on the states. We observed coordinated changes of histone mark signals at enhancers and promoters with gene expression. We detected evolutionary conserved bivalent promoters, selectively silencing ~3,400 genes, including hundreds of TFs regulating embryonic development. Second, I present a supervised method to predict TF binding across cell types, with features based on DNA sequence and patterns in DNase I cleavage data. We found that sequence and DNase read counts can outperform other features as well as state-of-the-art methods. I also describe our contribution to the ENCODE TF Binding DREAM challenge where we developed a method, using multiscale features and Extreme Boosting. Third, I describe methods, tools, and computational infrastructure that we have developed to handle large amounts of experimental data and metadata. These tools are fundamental to the selection and integration of large experimental datasets and are at the core of our pipelines, which are described in this dissertation. Finally, I present the protein docking server I developed, as well as algorithms and routines for post-processing predictions and protein structures. Collectively, this body of work encompasses computational approaches to the analyses of chromatin states, gene regulation, and the integration of large experimental datasets. / 2021-08-31T00:00:00Z

Molecular interaction among proteins drives the cellular processes through the formation of complexes that perform the requisite biochemical function. While some of the complexes are obligate (i.e., they fold together while complexation) others are non-obligate, and are formed through macromolecular recognition. Macromolecular recognition in proteins is highly specific, yet it can be both permanent and non permanent in nature. Hallmarks of permanent recognition complexes include large surface of interaction, stabilization by hydrophobic interaction and other noncovalent forces. Several amino acids which contribute critically to the free energy of binding at these interfaces are called as “hot spot” residues. The non permanent recognition complexes, on the other hand, usually show small interface of interaction, with limited stabilization from non covalent forces. For both the permanent and non permanent complexes, the specificity of molecular interaction is governed by the geometric compatibility of the interaction surface, and the noncovalent forces that anchor them. A great deal of studies has already been performed in understanding the basis of protein macromolecular recognition.1; 2 Based on these studies efforts have been made to develop protein-protein docking algorithms that can predict the geometric orientation of the interacting molecules from their individual unbound states. Despite advances in docking methodologies, several significant difficulties remain.1 Therefore, in this thesis, we start with literature review to understand the individual merits and demerits of the existing approaches (Chapter 1),3 and then, we attempt to address some of the problems by developing methods to infer protein quaternary structure from the crystalline state, and improve structural and chemical understanding of protein-protein interactions through biological complex prediction. The understanding of the interaction geometry is the first step in a protein-protein interaction study. Yet, no consistent method exists to assess the geometric compatibility of the interacting interface because of its highly rugged nature. This suggested that new sensitive measures and methods are needed to tackle the problem. We, therefore, developed two new and conceptually different measures using the Delaunay tessellation and interface slice selection to compute the surface complementarity and atom packing at the protein-protein interface (Chapter 2).4 We called these Normalized Surface Complementarity (NSc) and Normalized Interface Packing (NIP). We rigorously benchmarked the measures on the non redundant protein complexes available in the Protein Data Bank (PDB) and found that they efficiently segregate the biological protein-protein contacts from the non biological ones, especially those derived from X-ray crystallography. Sensitive surface packing/complementarity recognition algorithms are usually computationally expensive and thus limited in application to high-throughput screening. Therefore, special emphasis was given to make our measure compute-efficient as well. Our final evaluation showed that NSc, and NIP have very strong correlation among themselves, and with the interface area normalized values available from the Surface Complementarity program (CCP4 Suite: <http://smb.slac.stanford.edu/facilities/software/ccp4/html/sc.html>); but at a fraction of the computing cost. After building the geometry based surface complementarity and packing assessment methods to assess the rugged protein surface, we advanced our goal to determine the stabilities of the geometrically compatible interfaces formed. For doing so, we needed to survey the quaternary structure of proteins with various affinities. The emphasis on affinity arose due to its strong relationship with the permanent and non permanent life-time of the complex. We, therefore, set up data mining studies on two databases named PQS (Protein Quaternary structure database: http://pqs.ebi.ac.uk) and PISA (Protein Interfaces, Surfaces and Assemblies: www.ebi.ac.uk/pdbe/prot_int/pistart.html) that offered downloads on quaternary structure data on protein complexes derived from X-ray crystallographic methods. To our surprise, we found that above mentioned databases provided the valid quaternary structure mostly for moderate to strong affinity complexes. The limitation could be ascertained by browsing annotations from another curated database of protein quaternary structure (PiQSi:5 supfam.mrc-lmb.cam.ac.uk/elevy/piqsi/piqsi_home.cgi) and literature surveys. This necessitated that we at first develop a more robust method to infer quaternary structures of all affinity available from the PDB. We, therefore, developed a new scheme focused on covering all affinity category complexes, especially the weak/very weak ones, and heteromeric quaternary structures (Chapter 3).6 Our scheme combined the naïve Bayes classifier and point-group symmetry under a Boolean framework to detect all categories of protein quaternary structures in crystal lattice. We tested it on a standard benchmark consisting of 112 recognition heteromeric complexes, and obtained a correct recall in 95% cases, which are significantly better than 53% achieved by the PISA,7 a state-of-art quaternary structure detection method hosted at the European Bioinformatics Institute, Hinxton, UK. A few cases that failed correct detection through our scheme, offered interesting insights into the intriguing nature of protein contacts in the lattice. The findings have implications for accurate inference of quaternary states of proteins, especially weak affinity complexes, where biological protein contacts tend to be sacrificed for the energetically optimal ones that favor the formation/stabilization of the crystal lattice. We expect our method to be used widely by all researchers interested in protein quaternary structure and interaction. Having developed a method that allows us to sample all categories of quaternary structures in PDB, we set our goal in addressing the next problem that of accurately determining stabilities of the geometrically compatible protein surfaces involved in interaction. Reformulating the question in terms of protein-protein docking, we sought to ask how we could reliably infer the stabilities of any arbitrary interface that is formed when two protein molecules are brought sterically closer. In a real protein docking exercise this question is asked innumerable times during energy-based screening of thousands of decoys geometrically sampled (through rotation+translation) from the unbound subunits. The current docking methods face problems in two counts: (i), the number of interfaces from decoys to evaluate energies is rather large (64320 for a 9º rotation and translation for a dimeric complex), and (ii) the energy based screening is not quite efficient such that the decoys with native-like quaternary structure are rarely selected at high ranks. We addressed both the problems with interesting results. Intricate decoy filtering approaches have been developed, which are either applied during the search stage or the sampling stage, or both. For filtering, usually statistical information, such as 3D conservation information of the interfacial residues, or similar facts is used; more expensive approaches screen for orientation, shape complementarity and electrostatics. We developed an interface area based decoy filter for the sampling stage, exploiting an assumption that native-like decoys must have the largest, or close to the largest, interface (Chapter 4).8 Implementation of this assumption and standard benchmarking showed that in 91% of the cases, we could recover native-like decoys of bound and unbound binary docking-targets of both strong and weak affinity. This allowed us to propose that “native-like decoys must have the largest, or close to the largest, interface” can be used as a rule to exclude non native decoys efficiently during docking sampling. This rule can dramatically clip the needle-in-a-haystack problem faced in a docking study by reducing >95% of the decoy set available from sampling search. We incorporated the rule as a central part of our protein docking strategy. While addressing the question of energy based screening to rank the native-like decoys at high rank during docking, we came across a large volume of work already published. The mainstay of most of the energy based screenings that avoid statistical potential, involve some form of the Coulomb’s potential, Lennard Jones potential and solvation energy. Different flavors of the energy functions are used with diverse preferences and weights for individual terms. Interestingly, in all cases the energy functions were of the unnormalized form. Individual energy terms were simply added to arrive at a final score that was to be used for ranking. Proteins being large molecules, offer limited scope of applying semi-empirical or quantum mechanical methods for large scale evaluation of energy. We, therefore, developed a de novo empirical scoring function in the normalized form. As already stated, we found NSc and NIP to be highly discriminatory for segregating biological and non biological interface. We, therefore, incorporated them as parameters for our scoring function. Our data mining study revealed that there is a reasonable correlation of -0.73 between normalized solvation energy and normalized nonbonding energy (Coulombs + van der Waals) at the interface. Using the information, we extended our scoring function by combining the geometric measures and the normalized interaction energies. Tests on 30 unbound binary protein-protein complexes showed that in 16 cases we could identify at least one decoy in top three ranks with ≤10 Å backbone root-mean-square-deviation (RMSD) from true binding geometry. The scoring results were compared with other state-of-art methods, which returned inferior results. The salient feature of our scoring function was exclusion of any experiment guided restraints, evolutionary information, statistical propensities or modified interaction energy equations, commonly used by others. Tests on 118 less difficult bound binary protein-protein complexes with ≤35% sequence redundancy at the interface gave first rank in 77% cases, where the native like decoy was chosen among 1 in 10,000 and had ≤5 Å backbone RMSD from true geometry. The details about the scoring function, results and comparison with the other methods are extensively discussed in Chapter 5.9 The method has been implemented and made available for public use as a web server - PROBE (http://pallab.serc.iisc.ernet.in/probe). The development and use of PROBE has been elaborated in Chapter 7.10 On course of this work, we generated huge amounts of data, which is useful information that could be used by others, especially “protein dockers”. We, therefore, developed dockYard (http://pallab.serc.iisc.ernet.in/dockYard) - a repository for protein-protein docking decoys (Chapter 6).11 dockYard offers four categories of docking decoys derived from: Bound (native dimer co-crystallized), Unbound (individual subunits as well as the target are crystallized), Variants (match the previous two categories in at least one subunit with 100% sequence identity), and Interlogs (match the previous categories in at least one subunit with ≥90% or ≥50% sequence identity). There is facility for full or selective download based on search parameters. The portal also serves as a repository to modelers who may want to share their decoy sets with the community. In conclusion, although we made several contributions in development of algorithms for improved protein-protein docking and quaternary structure inference, a lot of challenges remain (Chapter 8). The principal challenge arises by considering proteins as flexible bodies, whose conformational states may change on quaternary structure formation. In addition, solvent plays a major role in the free energy of binding, but its exact contribution is not straightforward to estimate. Undoubtedly, the cost of computation is one of the limiting factors apart from good energy functions to evaluate the docking decoys. Therefore, the next generation of algorithms must focus on improved docking studies that realistically incorporate flexibility and solvent environment in all their evaluations.

Le développement des anticorps thérapeutiques s'est rapidement accéléré dans les 10 dernières années et concerne un nombre croissant de pathologies. La connaissance de l'épitope, à savoir la région de la cible à laquelle l'anticorps se fixe, est essentielle pour la compréhension des effets fonctionnels de ce dernier. Nous avons développé une méthode in silico, MAbTope, qui permet une prédiction précise de cet épitope, quand bien même aucune structure 3D de l'anticorps d’intérêt n'est résolue. Cette méthode se base sur une méthode d'amarrage protéine-protéine développée auparavant dans l’équipe BIOS. Le jeu d'apprentissage a été fortement enrichi en complexes anticorps-cibles, de nouvelles fonctions de score spécifiques ont été mises au point, et le plus important, l'objectif de l'apprentissage-machine a été modifié pour optimiser non plus la conformation de !'assemblage, mais la prédiction de l'épitope. Nous montrons que la méthode qui en résulte permet une prédiction précise et robuste de l'épitope, que la structure 3D de l'anticorps soit connue ou non. Nous montrons également comment les prédictions peuvent être facilement exploitées pour la validation expérimentale. Enfin, nous montrons comment la méthode peut être utilisée pour étudier à haut-débit le recouvrement d'épitopes par des anticorps ayant la même cible. / The development of therapeutic antibodies has been rapidly increasing in the last 10 years, with application to an increasing number of pathologies. The knowledge of the epitope, the region of the antigen to which the antibody binds, is crucial for understanding its functional effects. We have developed an in silico method, MAbTope, which allows the accurate prediction of the epitope, regardless of the availability of the 3D structure of the antibody of interest. This method is based on a protein-protein docking method previously developed in the BIOS group. The learning dataset was enlarged in antibody-antigen complexes, new specific scoring functions have been designed, and very importantly, the objective of machine-learning was switched from the conformational perspective towards the epitope determination perspective. We show that the resulting method allows robust and accurate prediction, whether or not the 3D structure of the antibody is available. We also show how the predictions can be easily exploited for experimental validation. Finally, we show how this method can be used for high-throughput epitope binning.

O uso de programas de computador para simular a formação de complexos entre proteínas é uma abordagem importante para melhor compreensão de como estas moléculas interagem. A representação paramétrica e a representação em polinômios tridimensionais de Zernike são ambas descrições compactas de superfícies moleculares baseadas em harmônicos esféricos utilizadas para visualização e comparação de superfícies moleculares e para atracamento de proteínas. Entretanto, apresentam limitações como restrição à topologia da superfície e dificuldade de representação de funções arbitrárias. Neste estudo, procurou-se refinar a capacidade de representação destes métodos para obtenção de elevada qualidade de reprodução e aplicabilidade a superfícies de topologia arbitrária. Através da análise de diversos algoritmos de suas etapas, foi possível identificar os estágios de cálculo de malha triangular de superfície molecular e de mapeamento esférico como os mais influentes na qualidade da representação paramétrica em harmônicos esféricos, e a alta sensibilidade a mudança de valores nas funções projetadas na qualidade da representação em Zernike 3D. A incorporação de um método de cálculo de superfícies que gera uma malha com elevada regularidade, aliado a um moderno algoritmo de mapeamento esférico garantiu baixo nível de distorções e obtenção de superfícies reconstruídas de alta qualidade. Uma técnica de detecção de similaridade entre superfícies de diferentes conformações de um ensemble permitiu compartilhamento de partes da descrição entre várias superfícies, com correspondente redução de volume de dados e de demanda por processamento, e com influência controlável nas distorções introduzidas. Um modo diferente de organização dos dados de entrada causou melhoria na qualidade de reconstrução de funções gerais em Zernike 3D, embora com a introdução de um mapa para restauração da função. Os resultados obtidos indicam aplicação promissora dos métodos em docking de proteínas com alto nível de detalhes. / Using computer programs to simulate protein complex formation is an important approach for better comprehension of the interaction mechanisms of such molecules. The parametric representation and the Zernike polynomial method are both compact representations of molecular surfaces based on spherical harmonics, used for visualization and comparison of molecules and for protein-protein docking. They pose, however, limitations regarding surface topology and difficulty in representing arbitrary functions. In this study, the representation capacity of such methods were refined to attain high quality reproductions and applicability to arbitrary topologies. Throughout the analysis of several algorithms, the stages of surface mesh calculation and spherical mapping were identified as highly influential on the quality of the spherical harmonics parametric representation, while high sensibility to changes in the function values were identified as an influential factor for projections in 3D Zernike. A surface calculation method that generates a highly regular mesh was adopted and paired with a modern spherical mapping algorithm to yield reconstructions with low level of distortions and high quality surfaces. Similarity among surfaces from different structures in a conformational ensemble were detected to allow sharing of portions of the description among several surfaces, with corresponding reduction in data volume and processing demand and controllable influence of distortions. A new input data organization method improved the reconstruction quality of general functions in 3D Zernike, although introducing a map to restore the function. Results indicate promising application of the methods in highly detailed protein-protein docking.

An in silico design algorithm has been developed to design binding ligands for protein targets of known three-dimensional structure. In this method, the binding energy of a candidate ligand is used to ascribe it a probability of binding. A sample of a virtual library of candidate ligands is then used to ascribe implicit weights to all the ligands in the library. These weights are used to obtain virtual sub-libraries which collectively carry a greater probability to bind to the target. This algorithm is presented along with validation studies on the different algorithmic components, demonstrating how optimization of the design method can be best achieved.

This dissertation proposes new optimization algorithms targeting protein-protein docking which is an important class of problems in computational structural biology. The ultimate goal of docking methods is to predict the 3-dimensional structure of a stable protein-protein complex. We study two specific problems encountered in predictive docking of proteins. The first problem is Side-Chain Positioning (SCP), a central component of homology modeling and computational protein docking methods. We formulate SCP as a Maximum Weighted Independent Set (MWIS) problem on an appropriately constructed graph. Our formulation also considers the significant special structure of proteins that SCP exhibits for docking. We develop an approximate algorithm that solves a relaxation of MWIS and employ randomized estimation heuristics to obtain high-quality feasible solutions to the problem. The algorithm is fully distributed and can be implemented on multi-processor architectures. Our computational results on a benchmark set of protein complexes show that the accuracy of our approximate MWIS-based algorithm predictions is comparable with the results achieved by a state-of-the-art method that finds an exact solution to SCP. The second problem we target in this work is protein docking refinement. We propose two different methods to solve the refinement problem. The first approach is based on a Monte Carlo-Minimization (MCM) search to optimize rigid-body and side-chain conformations for binding. In particular, we study the impact of optimally positioning the side-chains in the interface region between two proteins in the process of binding. We report computational results showing that incorporating side-chain flexibility in docking provides substantial improvement in the quality of docked predictions compared to the rigid-body approaches. Further, we demonstrate that the inclusion of unbound side-chain conformers in the side-chain search introduces significant improvement in the performance of the docking refinement protocols. In the second approach, we propose a novel stochastic optimization algorithm based on Subspace Semi-Definite programming-based Underestimation (SSDU), which aims to solve protein docking and protein structure prediction. SSDU is based on underestimating the binding energy function in a permissive subspace of the space of rigid-body motions. We apply Principal Component Analysis (PCA) to determine the permissive subspace and reduce the dimensionality of the conformational search space. We consider the general class of convex polynomial underestimators, and formulate the problem of finding such underestimators as a Semi-Definite Programming (SDP) problem. Using these underestimators, we perform a biased sampling in the vicinity of the conformational regions where the energy function is at its global minimum. Moreover, we develop an exploration procedure based on density-based clustering to detect the near-native regions even when there are many local minima residing far from each other. We also incorporate a Model Selection procedure into SSDU to pick a predictive conformation. Testing our algorithm over a benchmark of protein complexes indicates that SSDU substantially improves the quality of docking refinement compared with existing methods.

This thesis has been focused around the Diels Alder reaction with the goal to design an enzyme catalyzed reaction pathway. To achieve this goal computer aided enzyme design was utilized. Common traditional methods of computational chemistry (B3LYP, MP2) do not do well when calculating reaction barriers or even reaction energies for the Diels Alder reaction. New calcu- lation methods were developed and tested. This was the focus of the first part of the thesis, by choosing a small system, extensive and heavy calculations could be done with CBS-QB3. Then by benchmarking faster methods of calculation (SCS-MP2, M06-2X) against the results, they could be graded by efficiency and cost. This was done anticipating that the same accuracy could be applied to larger systems where CBS-QB3 cannot be used. In the second part activating groups were investigated for both the diene and the dienophile, along with their effects on reaction rates. A qualitative analysis was done. This is important not only for the uncatalyzed reaction, but also interesting when searching for possible substrates for the enzyme reaction. In the last part the thesis presents a designed enzyme that catalyzes Diels Alder in silico using ∆5−3−Keto steroid isomerase. Using empirical calculations, the enzyme was scanned for catalytic activity. The catalytic effect was then showed with ab initio Quantum chemical calculations.

Protein-protein and protein-peptide interactions play a central role in various aspects of the structural and functional organization of the cell. While the most complete structural characterization is provided by X-ray crystallography, many biological interactions occur in complexes that will not be amenable to direct experimental analysis. Therefore, it is important to develop computational docking methods that start from the structures of component proteins and predict the structure of their complexes, preferably with accuracy close to that provided by X-ray crystallography. This thesis details three applications of computational protein modeling, including the study of antibody maturation mechanisms, and the development of protocols for peptide-protein interaction prediction and template-based modeling of protein complexes. The first project, a comparative analysis of docking an antigen structure to antibodies across a lineage, reveals insights into antibody maturation mechanisms. A linear relationship between near-native docking results and changes in binding free energy is established, and used to investigate changes in binding affinity following mutation across two antibody-antigen systems: influenza and anthrax. The second project demonstrates that a motif-based search of available protein crystal structures is sufficient to adequately represent the conformational space sampled by a flexible peptide, compared to that of a rigid globular protein. This observation forms the basis for a global peptide-protein docking protocol that has since been implemented into the Structural Bioinformatics Laboratory’s docking web server, ClusPro. Finally, as structure availability remains a roadblock to many studies, researchers turn to homology modeling, in which the desired protein sequence is modeled onto a related structure. This is particularly challenging when the target is a protein complex, further restricting template availability. To address this problem, the third project details the development of a new template-based modeling protocol to be integrated into the ClusPro server. The implementation of a novel template-based search enables users to model both homomeric and heteromeric complexes, greatly expanding ClusPro server functionality. / 2020-08-30T00:00:00Z