EUVIMEDEUVIMED entdecken
Crossref

Graph Neural Network for 3‐Dimensional Structures Including Dihedral Angles for Molecular Property Prediction

Sri Abhirath Reddy Sangala, Shampa Raghunathan iD

Journal of Computational Chemistry · 15.05.2025 · 46 (13)

DOI: 10.1002/jcc.70121 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Computational Chemistry, online erschienen am 14.05.2025. Verfasst von Sri Abhirath Reddy Sangala und 1 weiteren Autor:innen. Beteiligte Einrichtung: École Centrale School of Engineering, Mahindra University, Hyderabad, India. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 56 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The prediction of molecular properties using graph neural network (GNN)‐ based approaches has attracted great attention in recent years. Topological molecular graphs are commonly used for representing molecules in machine learning (ML). However, the challenge is to utilize the complete geometry information, such as, bonds, angles, and dihedral angles while processing a molecular graph. In this work, we present predictive GNN accounting three‐dimensional molecular structures including the dihedral angles (GNN3Dihed) in a systematic manner. Additionally, we demonstrate that the usage of autoencoders to generate latent space embeddings for usually sparse atomic and bond vectors reduces the number of parameters in the message passing stage while not reducing performance. We compare the performance of GNN3Dihed with state‐of‐the‐art baselines on several tasks (regression and classification), for example, solubility prediction, toxicity prediction, binding affinity, and quantum mechanical property prediction, and showed that the present architecture often outperforms other models—demonstrating the importance of 3D structural information for ML in chemistry.

Schlagwörter

Stichwörter der Autor:innen

Quantum mechanical propertybinding affinitydipole momentgraph neural networkmachine learningmolecular property predictionregression and classification taskssolubilitythree‐dimensional structuretoxicity

Autor:innen und Institutionen

  • École Centrale School of Engineering, Mahindra University, Hyderabad, India.

ORCID-Profile: Shampa Raghunathan

Zitieren

APA

Sangala, S. A. R., & Raghunathan, S. (2025). Graph Neural Network for 3‐Dimensional Structures Including Dihedral Angles for Molecular Property Prediction. Journal of Computational Chemistry, 46(13). https://doi.org/10.1002/jcc.70121

Vancouver

Sangala SAR, Raghunathan S. Graph Neural Network for 3‐Dimensional Structures Including Dihedral Angles for Molecular Property Prediction. Journal of Computational Chemistry. 2025;46(13). doi:10.1002/jcc.70121

BibTeX
@article{Sangala2025,
  author = {Sangala, Sri Abhirath Reddy and Raghunathan, Shampa},
  title = {Graph Neural Network for 3‐Dimensional Structures Including Dihedral Angles for Molecular Property Prediction},
  journal = {Journal of Computational Chemistry},
  year = {2025},
  volume = {46},
  number = {13},
  doi = {10.1002/jcc.70121}
}
Kostenlos starten →