AET5: A transcriptome-guided molecular generation framework with contrastive self-supervised learning
8 affiliations
- 1School of Electronic and Information Engineering, Suzhou University of Science and Technology, Suzhou, Jiangsu, China
- 2Hangzhou Institute of Medicine, Chinese Academy of Sciences, Hangzhou, Zhejiang, China
- 3School of Computer Science and Technology, Zhejiang Normal University, Jinhua, Zhejiang, China
- 4Yangtze Delta Region Institute (Quzhou), University of Electronic Science and Technology of China, Quzhou, Zhejiang, China
- 5University of Oxford, Oxford, Oxfordshire, United Kingdom
- 6Faculty of Computing, Harbin Institute of Technology, Harbin, Heilongjiang, China
- 7School of Information Technology, Halmstad University, Halmstad, Sweden
- 8Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong SAR, China
Abstract
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. Author summary: Gene expression profiles provide a system-level view of how cells respond to diseases and drug treatments, making them valuable for computational drug discovery. However, directly using transcriptomic data to guide de novo molecular generation remains challenging because these data are often sparse, noisy, and difficult to map reliably to valid chemical structures. In this study, we propose AET5, an expression-guided molecular generation framework that designs molecules conditioned on disease-reversal gene expression profiles. AET5 combines contrastive self-supervised learning with a pre-trained sequence-to-sequence molecular generator to learn robust associations between transcriptomic signatures and molecular structures, thereby reducing the influence of experimental noise and improving the alignment between biological perturbation information and chemical space. Experiments on the L1000 dataset show that AET5 achieves improved overall generation performance compared with existing expression-guided methods and produces molecules with favorable structural, distributional, and physicochemical characteristics. Case studies on SARS-CoV-2 infection and prostate cancer further suggest that the generated compounds may have biologically relevant target-binding potential. Overall, AET5 provides a scalable and biologically informed strategy for transcriptome-guided molecular design.
Related
- Improving Molecular-Morphology Contrastive Pretraining using Deep-Learning-based Morphology Profiles
- scLAP: A Liver-Adapted Contrastive Shared-Latent Framework for Pseudo-Pairing Single-Cell Transcriptomes and Chromatin Accessibility Profiles
- meth-SemiCancer2: a cancer subtyping framework leveraging domain adaptation with contrastive learning for batch effect correction in DNA methylation profiles
- Full-length transcriptome and RNA-Seq analyses reveal the molecular mechanism of light protection in Hosta ventricosa.
- Molecular Safeguards of Survival: De novo transcriptome assembly and tissue-specific transcriptomic profiling of the yellow-foot clam Paphia malabarica
BibTeX
@article{yuan2026aet5,
title = {{AET5: A transcriptome-guided molecular generation framework with contrastive self-supervised learning}},
author = {Zhikang Yuan and Xin Zhang and Gaoming Lin and Quan Zou and Subhashisa Swain and Yijie Ding and Prayag Tiwari and Shuofeng Yuan and Xiaoyi Guo},
journal = {PLOS Computational Biology},
year = {2026},
month = sep,
doi = {10.1371/journal.pcbi.1014703},
url = {https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014703},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - AET5: A transcriptome-guided molecular generation framework with contrastive self-supervised learning
AU - Zhikang Yuan
AU - Xin Zhang
AU - Gaoming Lin
AU - Quan Zou
AU - Subhashisa Swain
AU - Yijie Ding
AU - Prayag Tiwari
AU - Shuofeng Yuan
AU - Xiaoyi Guo
PY - 2026
DA - 2026/09/10
JO - PLOS Computational Biology
DO - 10.1371/journal.pcbi.1014703
UR - https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014703
AB - 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. Author summary: Gene expression profiles provide a system-level view of how cells respond to diseases and drug treatments, making them valuable for computational drug discovery. However, directly using transcriptomic data to guide de novo molecular generation remains challenging because these data are often sparse, noisy, and difficult to map reliably to valid chemical structures. In this study, we propose AET5, an expression-guided molecular generation framework that designs molecules conditioned on disease-reversal gene expression profiles. AET5 combines contrastive self-supervised learning with a pre-trained sequence-to-sequence molecular generator to learn robust associations between transcriptomic signatures and molecular structures, thereby reducing the influence of experimental noise and improving the alignment between biological perturbation information and chemical space. Experiments on the L1000 dataset show that AET5 achieves improved overall generation performance compared with existing expression-guided methods and produces molecules with favorable structural, distributional, and physicochemical characteristics. Case studies on SARS-CoV-2 infection and prostate cancer further suggest that the generated compounds may have biologically relevant target-binding potential. Overall, AET5 provides a scalable and biologically informed strategy for transcriptome-guided molecular design.
SN - 1553-7358
ER -