Programme

P.87 -- Posters, VCTP-50

Date: Thursday, 7 August 2025

Time: 08:30 - 10:00

Enhance sampling of amyloid beta peptide: the combination of molecular dynamics simulations and machine learning

Pham Nguyen Tin Trung, Tran Thi Minh Thu

Faculty of Materials Science and Technology, Ho Chi Minh City University of Science VNUHCM, 227 Nguyen Van Cu Street, District 5, Ho Chi Minh City, Viet Nam Vietnam National University, Ho Chi Minh City 700000, Vietnam

Molecular dynamics (MD) simulations are powerful and widely used tools for studying protein structure and function. However, conventional MD simulations often require extensive computational time to sufficiently sample the conformational space. To address this limitation, various enhanced sampling techniques such as replica exchange, umbrella sampling, and metadynamics have been developed. In this study, we combine molecular dynamics simulations with machine learning to explore the conformational space of the amyloid beta protein, both in its full-length form and in a truncated segment (residues 18–28). Specifically, we employ an autoencoder to project MD-generated conformations onto a low-dimensional conformational landscape defined by principal component analysis (PCA). Our results demonstrate that this integrated approach can serve as a useful predictive tool for identifying low-energy conformations of proteins amyloid beta and offers a promising alternative for enhanced sampling in MD simulations.

Presenter: Tran Thi Minh Thu