Subject Area
Biochemistry, Chemistry, Computer Science
Abstract
The drug designing process is an intricate and resource-intensive endeavor, spanning approximately 12 years to transform a lead compound into a functional drug. Computer-aided drug design (CADD) plays a fundamental role in this process, mitigating the risk of failures in the later stage and saving valuable time and money. In the last decade, Artificial intelligence (AI) has provided remarkable alternatives for practical challenges, including drug design. Simultaneously, advanced QM calculations have achieved near-experimental accuracy in simulating drug-target interactions. However, introducing QM and AI-based techniques into drug discovery presents considerable challenges. AI techniques require high- quality data and often struggle with the explainability of their predictions. On the other hand, QM calculations, such as Hartree-Fock (HF) and Density Functional Theory (DFT), are significantly more expensive when screening large molecular databases. Over the past four years, I have worked on finding solutions to these issues. These challenges have been addressed by introducing a combined molecular dynamics and QM/MM model to study the NS3 and NS5 non-structural proteins in the Dengue virus, identifying three promis- ing lead compounds: Kaempferol, Chlorogenic acid, and Quercetin, constructing a QM-level protein-ligand interaction database to analyze interaction patterns, developing a high-quality DFT-level database, QM40, for machine learning purposes, and creating an explainability algorithm, XInsight, and finally developing SmartCADD, a drug discovery platform integrating QM and AI techniques. Through these advancements, we aim to revolutionize how we discover and develop therapeutic solutions, playing a pivotal role in tackling critical health challenges.
Degree Date
Summer 8-6-2024
Document Type
Dissertation
Degree Name
Ph.D.
Department
Department of Chemistry
Advisor
Elfi Kraka
Format
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License
Recommended Citation
Mahamada Kalapuwage, Ayesh Madushanka, "INTEGRATION OF MACHINE LEARNING AND QUANTUM MECHANICAL CALCULATIONS FOR DE NOVO DRUG DESIGN" (2024). Chemistry Theses and Dissertations. 47.
https://scholar.smu.edu/hum_sci_chemistry_etds/47
