EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

AET5: A transcriptome-guided molecular generation framework with contrastive self-supervised learning

Zhikang Yuan, Xin Zhang iD, Gaoming Lin, Quan Zou iD, Subhashisa Swain iD, Yijie Ding iD, Prayag Tiwari, Shuofeng Yuan, Xiaoyi Guo

PLOS Computational Biology · 2026

DOI: 10.1371/journal.pcbi.1014703 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS Computational Biology, 2026. Verfasst von Zhikang Yuan und 8 weiteren Autor:innen. Beteiligte Einrichtung: School of Electronic and Information Engineering, Suzhou University of Science and Technology, Suzhou, Jiangsu, China. Hauptthemen laut MEDLINE: Supervised Machine Learning, Transcriptome, Gene Expression Profiling. Die Arbeit nennt 68 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Municipal Government of Quzhou, Guangdong-Hong Kong Technology Cooperation, Government of the Hong Kong Special Administrative Region.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Gene expression profiles capture system-level drug responses and offer a promising basis for de novo molecular generation. However, their application is limited by data sparsity and experimental noise, which hinder the reliable mapping between disease-associated transcriptomic perturbations and chemically valid therapeutic molecules. Here, we present AET5, a de novo molecular generation framework that conditions molecular design on disease-reversal gene expression profiles. AET5 integrates contrastive self-supervised learning with pre-trained sequence-to-sequence models to learn robust associations between transcriptomic signatures and molecular structures by deriving noise-tolerant transcriptomic representations and aligning them with molecular sequence space. Across the L1000 dataset, AET5 outperforms existing expression-guided generation methods in generation quality and distributional characteristics, while maintaining favorable physicochemical and drug-related properties. We further apply AET5 to generate candidate compounds for SARS-CoV-2 infection and prostate cancer. Molecular docking and dynamics simulations indicate stable target binding, supporting the biological relevance of the generated molecules. These results demonstrate that disease-reversal expression profiles can effectively guide de novo molecular generation, providing a general framework for biologically informed drug design under noisy transcriptomic conditions.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Supervised Machine LearningTranscriptomeGene Expression ProfilingHumansProstatic NeoplasmsMolecular Docking SimulationComputational BiologySARS-CoV-2Molecular Dynamics SimulationCOVID-19 Drug TreatmentCOVID-19

Autor:innen und Institutionen

  • School of Electronic and Information Engineering, Suzhou University of Science and Technology, Suzhou, Jiangsu, China.
  • Hangzhou Institute of Medicine, Chinese Academy of Sciences, Hangzhou, Zhejiang, China.
  • School of Computer Science and Technology, Zhejiang Normal University, Jinhua, Zhejiang, China.
  • Yangtze Delta Region Institute (Quzhou), University of Electronic Science and Technology of China, Quzhou, Zhejiang, China.
  • University of Oxford, Oxford, Oxfordshire, United Kingdom.
  • Faculty of Computing, Harbin Institute of Technology, Harbin, Heilongjiang, China.

ORCID-Profile: Xin Zhang, Quan Zou, Subhashisa Swain, Yijie Ding

Förderung

  • Municipal Government of Quzhou
  • Guangdong-Hong Kong Technology Cooperation
  • Government of the Hong Kong Special Administrative Region
  • the Municipal Government of Quzhou
  • Hong Kong Government

Zitieren

APA

Yuan, Z., Zhang, X., Lin, G., Zou, Q., Swain, S., Ding, Y., Tiwari, P., Yuan, S., & Guo, X. (2026). AET5: A transcriptome-guided molecular generation framework with contrastive self-supervised learning. PLOS Computational Biology. https://doi.org/10.1371/journal.pcbi.1014703

Vancouver

Yuan Z, Zhang X, Lin G, Zou Q, Swain S, Ding Y, et al. AET5: A transcriptome-guided molecular generation framework with contrastive self-supervised learning. PLOS Computational Biology. 2026. doi:10.1371/journal.pcbi.1014703

BibTeX
@article{Yuan2026,
  author = {Yuan, Zhikang and Zhang, Xin and Lin, Gaoming and Zou, Quan and Swain, Subhashisa and Ding, Yijie and Tiwari, Prayag and Yuan, Shuofeng and Guo, Xiaoyi},
  title = {AET5: A transcriptome-guided molecular generation framework with contrastive self-supervised learning},
  journal = {PLOS Computational Biology},
  year = {2026},
  doi = {10.1371/journal.pcbi.1014703}
}
Kostenlos starten →