Browsing by Author "Paton, Robert, advisor"

Now showing 1 - 3 of 3

Embargo
Data-driven strategies for organic structure-property and structure-reactivity relationships
(Colorado State University. Libraries, 2024) Santhanalakkshmi Vejaykummar, Shree Sowndarya, author; Paton, Robert, advisor; Prasad, Ashok, committee member; Kim, Seonah, committee member; Nielsen, Aaron, committee member
The prediction of molecular properties plays a pivotal role in various domains, from drug discovery to materials science. With the advent of machine learning (ML) techniques, particularly in the field of cheminformatics, the prediction of properties for small organic molecules has witnessed significant advancements. This document delves into the diverse machine-learning strategies employed for the accurate prediction of properties crucial for understanding molecular behavior. In Chapter 1, I offer insights into the evolution of data-driven modeling through Quantitative Structure-Property Relationships (QSPR), highlighting promising advancements in utilizing chemical features to construct predictive models for molecular properties. In Chapter 2, I delve into the primary stage of modeling, focusing on data collection for predictive tasks. I illustrate how the integration of automation and computational tools' advancement can construct modular workflows for FAIR (Findable, Accessible, Interoperable, and Reusable) chemistry. This approach aims to enhance the usability and reproducibility of scientific data. In Chapter 3, I emphasize leveraging computational tools to access high-level data for small organic molecules. I showcase the creation of a novel metric for assessing organic radical stability, utilizing a comprehensive chemical database of radicals. This involves employing straightforward physical organic descriptors, namely fractional spin, and buried volume, computed through systematic computational workflows. In Chapter 4, I explore the progression of graph-based models designed to forecast molecular properties, specifically Bond Dissociation Energy. Additionally, I conduct a thorough examination of two particular applications pertinent to pharmaceutical and atmospheric chemistry. I demonstrate that utilizing a minimal number of molecules from the relevant chemical space can notably enhance large-scale machine-learning models. Finally, in Chapter 5, I combine the developed tools from Chapters 3 and 4, to perform goal-directed molecular optimization in identifying novel radicals for aqueous redox flow batteries using graph neural networks (radical stability, redox potentials, and bond dissociation energy) and reinforcement learning. This de novo molecular optimization strategy has successfully identified 32 new radical candidates. By amalgamating insights from diverse studies, this dissertation endeavors to offer a comprehensive grasp of how machine-learning strategies are transforming the terrain of molecular property prediction.
Open Access
Modeling conformational heterogeneity in biomolecules
(Colorado State University. Libraries, 2023) Klem, Heidi, author; Paton, Robert, advisor; McCullagh, Martin, advisor; Levinger, Nancy, committee member; Kennan, Alan, committee member; Geiss, Brian, committee member
Regulation of biocatalytic cascades is essential for biological processes but has yet to be exploited in real-world applications. Allostery is a prime example, where binding of an effector molecule alters function in a remote location of the same biomolecule. V-type allostery is especially fascinating, as the reaction rate can be either increased or decreased in response to effector binding. Determining how conformational changes affect the reaction rate is challenging due to the disparity of timescales between the underlying molecular processes. Experimental methods, such as X-ray crystallography, can help to capture large-scale conformational change. However, the resulting structures are not guaranteed to correspond to the biophysical state relevant to the research questions being addressed. Structural changes that occur during the chemical reaction are particularly elusive to this approach. To understand the connection between conformational change and catalytic consequence, a description of the reaction mechanism and relevant configurations is needed. Quantum mechanical (QM) methods can be used to propose enzyme reaction mechanisms by modeling femtosecond motions of forming and breaking bonds. Large-scale conformational changes take place over much longer timescales that cannot be simulated at the QM level, therefore requiring classical simulation techniques. This dissertation focuses on the challenges posed by conformational change in the field of computational biocatalysis. The first chapter examines the prevalence of conformational change in enzymes, its relationship to catalysis, and the difficulties it presents. The second chapter looks at the influence of active site structural features on reaction rates in the allosteric enzyme IGPS using QM approaches and energy decomposition schemes. The third chapter covers the development of methods that use molecular dynamics (MD) simulations to analyze relevant structural states from simulation data and identify long-range communication pathways in biomolecules. The fourth chapter presents a Python code, enzyASM, that automates the generation of QM-based truncated active site models and discusses ongoing developments that will aid reproducibility and standardization in this field of research. The fifth and final chapter summarizes the implications of this Thesis work in computational biocatalysis and envisions how remaining challenges can be addressed to maximize potential to solve real-world problems.
Open Access
Understanding selectivity in organic reactions through density functional theory
(Colorado State University. Libraries, 2024) de Lescure, Louis Raymond Philibert, author; Paton, Robert, advisor; McNally, Andrew, advisor; Bandar, Jeffrey, committee member; Kennan, Alan, committee member; Herrera-Alonso, Margarita, committee member
The success of chemical reactions is often expressed through the lens of selectivity, defined as the preference for a desired reaction pathway over an undesirable one. A profound understanding of the rationale behind the selectivity of chemical reactions is crucial for the progression of synthetic methodologies in organic chemistry. Utilizing quantum chemical approximations, density functional theory (DFT) calculations offer unparalleled insights into the electronic structures and mechanisms of reactions, which can be correlated with observed empirical selectivities. This dissertation demonstrates the significant utility of DFT, in tandem with experimental evidence, in elucidating the intricate mechanisms of reactions. Chapter 1 defines the thematic and methods used throughout this thesis. Chapters 2 and 3 detail collaborative work with the McNally group at Colorado State University. Here, we developed innovative methods for the halogenation of pyridines and advanced modifications of pyrimidine rings utilizing redesigned Zincke chemistry. This chapter focuses on the factors influencing the regioselectivity of halogenation processes and provides mechanistic insights into the formation of crucial intermediates. Chapter 3 outlines a joint project with the Race group at the University of Minnesota, where we explored the homologation of benzylic carbon-bromide bonds. Our investigations centered on the ring-opening of phenonium intermediates, a critical step in determining the success of the reaction. Chapter 4 presents a collaboration with the Aggarwal group at the University of Bristol. This chapter examines the nuanced interplay between kinetic and thermodynamic factors that govern the enantioselectivity of the reaction discussed. This comprehensive study underscores the integration of theoretical and experimental approaches in advancing our understanding of complex chemical reactions.