EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multimodal Contrast-Enhanced Molecular Representation Learning and Property Prediction

Hong Luo, Jie He, Zhichao Liu, Chen Zeng

Biophysica · 27.03.2026 · 6 (2) · 24

DOI: 10.3390/biophysica6020024 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biophysica, online erschienen am 27.03.2026. Verfasst von Hong Luo und 3 weiteren Autor:innen. Die Arbeit nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, China Postdoctoral Science Foundation, Scientific and Technological Research Program of Chongqing Municipal Education Commission.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Molecular representation learning (MRL) has garnered significant attention due to its pivotal role in downstream applications such as molecular property prediction and drug discovery. In most MRL approaches, molecules are encoded into 2D topological graphs via graph neural network (GNN), which suffers from over-smoothing issues and limited receptive fields. Furthermore, most GNN models fail to utilize the 3D spatial structural information that determines molecular physicochemical properties and biological activity. To this end, here we propose multimodal contrast-enhanced molecular representation learning (MCMRL). This approach utilizes both the 2D topological information and 3D structural information of molecules for contrastive learning to enhance molecular graph representations. Further, it integrates additional molecular fingerprint information and feature fusion techniques to incorporate multimodal knowledge, yielding more reliable and generalizable molecular representations. MCMRL is pre-trained on ~10 million unlabeled molecules from PubChem, followed by various downstream benchmark tasks. Experimental results demonstrate that MCMRL achieves superior performance in 9 out of 13 benchmark tests for molecular property prediction, validating its effectiveness in molecular representation learning. Furthermore, potential molecular drugs binding to biological target protein DRD2 screened by MCMRL representation show promising affinity score, which also demonstrates the efficacy of the proposed method.

Förderung

  • National Natural Science Foundation of China
  • China Postdoctoral Science Foundation
  • Scientific and Technological Research Program of Chongqing Municipal Education Commission

Zitieren

APA

Luo, H., He, J., Liu, Z., & Zeng, C. (2026). Multimodal Contrast-Enhanced Molecular Representation Learning and Property Prediction. Biophysica, 6(2), 24. https://doi.org/10.3390/biophysica6020024

Vancouver

Luo H, He J, Liu Z, Zeng C. Multimodal Contrast-Enhanced Molecular Representation Learning and Property Prediction. Biophysica. 2026;6(2):24. doi:10.3390/biophysica6020024

BibTeX
@article{Luo2026,
  author = {Luo, Hong and He, Jie and Liu, Zhichao and Zeng, Chen},
  title = {Multimodal Contrast-Enhanced Molecular Representation Learning and Property Prediction},
  journal = {Biophysica},
  year = {2026},
  volume = {6},
  number = {2},
  pages = {24},
  doi = {10.3390/biophysica6020024}
}
Kostenlos starten →