CaltechTHESIS
  A Caltech Library Service

Physics-Informed Neural Approaches for Multiscale Molecular Modeling and Design

Citation

Qiao, Zhuoran (2023) Physics-Informed Neural Approaches for Multiscale Molecular Modeling and Design. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/48d1-ja21. https://resolver.caltech.edu/CaltechTHESIS:12102022-055022458

Abstract

Chemical processes in nature span multiple characteristic length and time scales, and the computational simulation for systems at the intersection of different scales is highly challenging with far-reaching implications for numerous scientific and industrial problems. To facilitate the computational modeling and design for large molecular systems and address the cost-resolution tradeoffs in conventional strategies, in this dissertation we introduce a series of physics-informed machine learning methods for the efficient computational modeling of chemical systems and the accurate prediction of their properties such as energetics, structures, and dynamics. In Chapters 2-3, we introduce a family of orbital-based geometric deep learning methods for the prediction of quantum chemical properties while adhering to the scaling and symmetry constraints of electronic structure theory. The presented methods achieve a chemical accuracy on community-wide benchmarks for molecular property prediction, and are shown to be transferable among diverse main-group molecular systems. In Chapter 4, we introduce a method for the prediction of protein-ligand complex structures based on a finite-time stochastic process parameterized by deep equivariant neural networks. The presented method achieves improved structure prediction accuracy against existing approaches, and is able to rapidly sample protein structures for folding landscapes that are modulated by inter-molecular interactions.

Item Type:Thesis (Dissertation (Ph.D.))
Subject Keywords:Molecular Simulation, Machine Learning, Electronic Structure, Generative Models
Degree Grantor:California Institute of Technology
Division:Chemistry and Chemical Engineering
Major Option:Chemistry
Minor Option:Applied Physics
Awards:Amazon AI4Science Fellowship, 2021
Thesis Availability:Public (worldwide access)
Research Advisor(s):
  • Anandkumar, Anima
Thesis Committee:
  • Goddard, William A., III (chair)
  • Anandkumar, Anima
  • Miller, Thomas F.
  • Wei, Lu
Defense Date:22 November 2022
Non-Caltech Author Email:zrqiao0 (AT) gmail.com
Funders:
Funding AgencyGrant Number
CaltechUNSPECIFIED
DeLogi Science and Technology GrantUNSPECIFIED
Amazon AI4Science FellowshipUNSPECIFIED
Record Number:CaltechTHESIS:12102022-055022458
Persistent URL:https://resolver.caltech.edu/CaltechTHESIS:12102022-055022458
DOI:10.7907/48d1-ja21
Related URLs:
URLURL TypeDescription
https://zrqiao.github.io/AuthorPersonal website
https://aip.scitation.org/doi/abs/10.1063/5.0021955PublisherArticle adapted for Chapter 2
https://arxiv.org/abs/2011.02680arXivArticle adapted for Chapter 2
https://www.pnas.org/doi/abs/10.1073/pnas.2205221119PublisherArticle adapted for Chapter 3
https://arxiv.org/abs/2209.15171arXivArticle adapted for Chapter 4
ORCID:
AuthorORCID
Qiao, Zhuoran0000-0002-5704-7331
Default Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:15077
Collection:CaltechTHESIS
Deposited By: Zhuoran Qiao
Deposited On:21 Dec 2022 16:21
Last Modified:08 Nov 2023 00:07

Thesis Files

[img] PDF - Final Version
Creative Commons Attribution Non-commercial Share Alike.

14MB

Repository Staff Only: item control page