Exploring Protein Conformations and Functions Through Molecular Dynamics Simulations and Machine Learning
Proteins are essential biomacromolecules that perform a variety of critical functions in living organisms. The tertiary structure of a protein plays a crucial role in its biological activity as it determines how the protein interacts with other molecules. Consequently, understanding protein conformation and function is an important area of research with implications in medicine, biotechnology, and other fields.
The first part of this dissertation focuses on protein allostery, a process by which proteins transmit perturbations caused by binding at one site to a distal site, thereby regulating activity. With the development of computational methods like molecular dynamics simulations and machine learning, it is now possible to study protein allostery at the atomistic level. A machine learning-based framework is presented to understand the allosteric process of the light-oxygen-voltage domain of the diatom Phaeodactylum tricornutum aureochrome 1a protein. Upon understanding allostery mechanisms, the identification of allosteric sites is of great importance in allosteric drug discovery and design. To approach this problem, the Protein Allosteric Sites Server (PASSer) is designed for accurate allosteric site prediction.
The second part of this dissertation explores protein conformations using deep learning. Variational autoencoders, a class of deep learning models, are utilized to learn a low-dimensional representation that captures the essential features of high-dimensional protein conformations. The success of this approach is demonstrated through the study of the enzyme adenosine kinase and Vivid. Furthermore, an adaptive sampling method is presented that can accelerate the exploration of protein conformational space.
Number of Pages
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License
Tian, Hao, "Exploring Protein Conformations and Functions Through Molecular Dynamics Simulations and Machine Learning" (2023). Chemistry Theses and Dissertations. 34.
Available for download on Thursday, April 25, 2024