A. James Clark School of Engineering
Permanent URI for this communityhttp://hdl.handle.net/1903/1654
The collections in this community comprise faculty research works, as well as graduate theses and dissertations.
Browse
2 results
Search Results
Item Molecular dynamics simulation and machine learning study of biological processes(2022) Ghorbani, Mahdi; Klauda, Jeffery B; Brooks, Bernard R; Chemical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)In this dissertation, I use computational techniques especially molecular dynamics (MD) and machine learning to study important biological processes. MD simulations can effectively be used to understand and investigate biologically relevant systems with lengths and timescales that are otherwise inaccessible to experimental techniques. These include but are not limited to thermodynamics and kinetics of protein folding, protein-ligand binding free energies, interaction of proteins with membranes, and designing new therapeutics for diseases with rational design strategies. The first chapter includes a detailed description of the computational methods including MD, Markov state modeling and deep learning. In the second chapter, we studied membrane active peptides using MD simulation and machine learning. Two cell penetrating peptides MPG and Hst5 were simulated in the presence of membrane. We showed that MPG enters the model membrane through its N-terminal hydrophobic residues while Hst5 remains attached to the phosphate layer. Formation of helical conformation for MPG helps its deeper insertion into membrane. Natural language processing (NLP) and deep generative modeling using a variational attention based variational autoencoder (VAE) was used to generate novel antimicrobial peptides. These in silico generated peptides have a high quality with similar physicochemical properties to real antimicrobial peptides. In the third chapter, we studied kinetics of protein folding using Markov state models and machine learning. We studied the kinetics of misfolding in β2-microglobulin using MSM analysis which gave us insights about the metastable states of β2m where the outer strands are unfolded and the hydrophobic core gets exposed to solvent and is highly amyloidogenic. In the next part of this chapter, we propose a machine learning model Gaussian mixture variational autoencoder (GMVAE) for simultaneous dimensionality reduction and clustering of MD simulations. The last part of this chapter is about a novel machine learning model GraphVAMPNet which uses graph neural networks and variational approach to markov processes for kinetic modeling of protein folding. In the last chapter, we study two membrane proteins, spike protein of SARS-COV-2 and EAG potassium channel using MD simulations. Binding free energy calculations using MMPBSA showed a higher binding affinity of receptor binding domain in SARS-COV-2 to its receptor ACE2 than SARS-COV which is one of the major reason for its higher infection rate. Hotspots of interaction were also identified at the interface. Glycans on the spike protein shield the spike from antibodies. Our MD simulation on the full length spike showed that glycan dynamics gives the spike protein an effective shield. However, breaches were found in the RBD at the open state for therapeutics using network analysis. In the last section, we study ligand binding to the PAS domain of EAG potassium channel and show that a residue Tyr71 blocks the binding pocket. Ligand binding inhibits the current through EAG channel.Item Examining the role of water and hydrophobicity in folding, aggregation, and allostery(2018) Custer, Gregory Scott; Matysiak, Silvina; Bioengineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Solvation and hydrophobicity drive many critical processes in nature, playing an important role in the folding of proteins, aggregation of surfactants into micelles, and in the disorder to order transitions that occur in some allosteric proteins upon ligand binding. Understanding how solvation and hydrophobicity affect these processes at a molecular level is important to finding new ways to use these processes, but it can be difficult to characterize these molecular details using experimental methods. Molecular dynamics (MD) simulations have proven useful in exploring details and thermodynamic conditions inaccessible in experiment, as MD captures the time evolution of the system at a molecular level. The phenomena which can be studied with an MD simulation depend on the mathematical model employed. Atomistic models provide the most detail for a simulation, but due to the computational costs required are not typically used to study phenomena which require large systems and time scales greater than several μs. Coarse-grained (CG) models reduce the complexity of the system being studied, enabling the exploration of phenomena that occur at longer time scales. We have developed CG models to study protein folding and surfactant aggregation. Our CG surfactant model uses a three-body potential to account for hydrogen bonding without an explicit electrostatic potential, reducing the computational cost of the model. With our surfactant model we studied the stability of non-ionic micelles at extremes of temperature, capturing a window of thermal stability with destabilization of the micelles at both high and low temperatures. We observed changes in structure and solvation of the micelle at low temperatures, with a shift in enthalpy of solvation water providing the driving force for destabilization. Solvation and hydrophobicity are also critical in the folding and stability of proteins. With a modified version of our surfactant model we characterized the folding landscape of a designed sequence which folds to a helical bundle in water. We found two competing folded states which differ by rotation of a helix and trade between hydrophobic packing and solvation of protein's core. Changes in hydrophobic packing can also be involved in the disorder to order transitions that occur upon liganding binding in an allosteric protein, such as the E. Coli biotin ligase/repressor (BirA), in which ligand binding promotes dimerization. We have used atomistic simulations of BirA mutants in collaboration with an experimental group to identify structural changes, accompanied by changes in solvation, at both the dimer interface and ligand binding regions for distal mutations which impact the functionality of BirA.