EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Enhancing cross-context generalization in drug perturbation prediction with a multimodal conditional diffusion framework

Yanjie Ma, Kang Du, Yan Li, Pengyong Li iD, Liang Yu iD

Bioinformatics · 02.07.2026 · 42 (7)

DOI: 10.1093/bioinformatics/btag482 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 30.06.2026. Verfasst von Yanjie Ma und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science and Technology, Xidian University, Xi'an, 710126, Shaanxi, China. Hauptthemen laut MEDLINE: Computational Biology, Transcriptome. Die Arbeit nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motivation Predicting drug-induced transcriptional perturbations is critical for precision medicine, yet existing models fail to capture multimodal biological context, limiting generalization across unseen drugs and cell lines. Results We present PertDiff, a conditional diffusion framework that integrates control gene expression, LLM-derived cell semantics, and pretrained molecular graph representations to predict transcriptome-wide perturbations. PertDiff outperforms state-of-the-art baselines in prediction accuracy and generalizes robustly across drugs and cell lines. It further demonstrates translational utility through accurate drug sensitivity prediction, therapeutic repurposing for pancreatic cancer, and concordance with real-world clinical treatment outcomes, establishing it as a biologically grounded transcriptomic modeling tool. Availability The source code and data are available at https://github.com/Panda-myj/PertDiff and https://doi.org/10.5281/zenodo.18427848.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Computational BiologyTranscriptomeHumansGene Expression ProfilingPrecision MedicinePancreatic NeoplasmsAntineoplastic Agents

Substanzen

Antineoplastic Agents

Autor:innen und Institutionen

  • School of Computer Science and Technology, Xidian University, Xi'an, 710126, Shaanxi, China.
  • School of Management, Xi'an Polytechnic University, Xi'an, 710000, Shaanxi, China.

ORCID-Profile: Pengyong Li, Liang Yu

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Ma, Y., Du, K., Li, Y., Li, P., & Yu, L. (2026). Enhancing cross-context generalization in drug perturbation prediction with a multimodal conditional diffusion framework. Bioinformatics, 42(7). https://doi.org/10.1093/bioinformatics/btag482

Vancouver

Ma Y, Du K, Li Y, Li P, Yu L. Enhancing cross-context generalization in drug perturbation prediction with a multimodal conditional diffusion framework. Bioinformatics. 2026;42(7). doi:10.1093/bioinformatics/btag482

BibTeX
@article{Ma2026,
  author = {Ma, Yanjie and Du, Kang and Li, Yan and Li, Pengyong and Yu, Liang},
  title = {Enhancing cross-context generalization in drug perturbation prediction with a multimodal conditional diffusion framework},
  journal = {Bioinformatics},
  year = {2026},
  volume = {42},
  number = {7},
  doi = {10.1093/bioinformatics/btag482}
}
Kostenlos starten →