Machine learning has boosted the remarkable development of crystal structure prediction (CSP), greatly accelerating modern materials design. However, slow location of the low-energy regions on the potential energy surface (PES) is still a key bottleneck for the overall search efficiency. Here, we develop a low-energy region explorer (LoreX) to rapidly locate low-energy regions on the PES. This achievement stems from graph-deep-learning-based PES slicing, which classifies structures into different prototypes to divide and conquer the PES. The accuracy and efficiency of LoreX are validated on 27 typical compounds, showing that it correctly locates low-energy regions with only 100 selected samples. The powerful capability of LoreX is demonstrated in solving two challenging problems: discovering new boron allotropes and identifying the puzzling crystal structures of the ordered vacancy compound CuInSe. This study establishes a new method for rapid PES exploration and offers a highly efficient and generally applicable approach to accelerating CSP.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1021/jacs.4c17343 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Associations of Device-Measured Physical Activity and Sedentary Time With Neural Responses to Visual Food Cues in Adults: A Functional Magnetic Resonance Imaging Study.
Hum Brain Mapp
March 2025
National Centre for Sport and Exercise Medicine, School of Sport, Exercise and Health Sciences, Loughborough University, Loughborough, UK.
Self-reported physical activity is associated with lower brain food cue responsiveness in reward-related regions, but relationships utilizing objective physical activity measurement tools have not been explored. This cross-sectional study examined whether device-measured moderate-to-vigorous intensity physical activity and sedentary time are related to neural responses to visual food cues using functional magnetic resonance imaging. Fifty-one healthy adults (30 men, 21 women; mean ± SD: age 26 ± 6 years; body mass index 24.
View Article and Find Full Text PDFLoreX: A Low-Energy Region Explorer Boosts Efficient Crystal Structure Prediction.
J Am Chem Soc
March 2025
Eastern Institute of Technology, Ningbo 315200, China.
Machine learning has boosted the remarkable development of crystal structure prediction (CSP), greatly accelerating modern materials design. However, slow location of the low-energy regions on the potential energy surface (PES) is still a key bottleneck for the overall search efficiency. Here, we develop a low-energy region explorer (LoreX) to rapidly locate low-energy regions on the PES.
View Article and Find Full Text PDFMyofascial pain syndrome (MPS) is a chronic musculoskeletal pain characterized by myofascial trigger points (MTrPs), which are related to dysfunctional motor endplate. However, the changes in neuromuscular functions incurred by the MTrPs are still not well understood. High-density surface electromyography (HD-sEMG) provides a non-invasive method to identify motor endplates and quantify the muscle activity distribution.
View Article and Find Full Text PDFDeepConf: Leveraging ANI-ML Potentials for Exploring Local Minima with Application to Bioactive Conformations.
J Chem Inf Model
March 2025
Department of Chemistry, Gebze Technical University, 41400 Kocaeli, Turkey.
Here, we introduce a low energy conformer generation algorithm using ANI-ML potentials at the DFT accuracy and benchmark in reproducing bioactive conformations. We show that the method is efficient when the initial starting structure is far from equilibrium, when the ML potentials are stuck in nonsmooth regions, and when the quality of the conformers in a less conformer size is demanded. We specifically focus on conformations due to rotations around the single bonds.
View Article and Find Full Text PDFMaximising the potential of type 2 diabetes remission: scale up and sustainability considerations from the DiRECT-Aus implementation trial.
Aust J Prim Health
March 2025
School of Primary and Allied Health Care, Monash University, Melbourne, Vic 3199, Australia.
Background Approximately 500 million people worldwide live with type 2 diabetes mellitus. The UK's 'Diabetes Remission Clinical Trial' (DiRECT) is a potential novel method for care. An Australian trial of DiRECT (DiRECT-Aus) showed that 56% of participants achieved diabetes remission at 12months.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!