Computational insights into biomolecular systems using artificial intelligence

In this thesis, the transformative potential of artificial intelligence (AI) in molecular sciences is explored, spanning protein dynamics, biological membrane behavior, polymer morphology prediction, novel material design, and drug development. Leveraging AI algorithms such as variational autoencoders (VAE) and deep neural networks (DNN), novel insights into the transition pathways of transmembrane transporter proteins are uncovered, alongside a deeper understanding of membrane properties through the integration of topological data analysis(TDA) with AI models. Additionally, the development of computational frameworks, including GHP-MOFassemble for accelerated discovery of metal-organic frameworks (MOFs) optimized for CO2 capture, and machine learning approaches for enhanced prediction of polymer morphology, demonstrates the transformative potential of AI in material design. Furthermore, AI methodologies in drug discovery facilitate the design of kinase inhibitor drugs with improved properties and selectivity, while the optimization of large AI models for protein structure prediction expedites drug discovery efforts. Through this interdisciplinary investigation, AI emerges as a powerful tool for addressing complex biological challenges and shaping the future of molecular sciences.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Biophysics & Quant Biology

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2024

Author dc:creator

• Park, Hyun

Contributors dc:contributor

• Tajkhorshid, Emad
• Aksimentiev, Aleksei
• Pogorelov, Taras
• Huerta, Eliu A.

• Ai
• Machine Learning
• Deep Learning
• Biophysics
• Biomolecule
• Protein
• Lipid
• Drug
• Polymer
• Mof
• Ai Framework
• Transition Pathway
• Generative Ai
• Drug Discovery
• Membrane
• Topological Data Analysis
• Alphafold
• Membtda
• Prottransvae
• Apace
• Ghpmof-assemble
• Diffusion Model
• Vae
• Predictive Model
• Molecular Dynamics
• Monte Carlo
• High Performance Computing
• Structure Prediction
• Conformational Diversity

Statement dc:rights

• Copyright 2024 Hyun Park

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/124634

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F124634