EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Generative modeling for RNA splicing prediction and design

Di Wu iD, Natalie Maus, Anupama Jha, Kevin Yang, Benjamin D Wales-McGrath, San Jewell, Anna Tangiyan, Peter Choi, Jake R Gardner, Yoseph Barash iD

eLife · 2026

DOI: 10.7554/elife.106043 ↗

Kurz gefasst

Fachartikel in der Zeitschrift eLife, 2026. Verfasst von Di Wu und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer and Information Science, School of Engineering, University of Pennsylvania, Philadelphia, United States. Hauptthemen laut MEDLINE: RNA Splicing, Alternative Splicing, Computational Biology. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 91 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NLM NIH HHS, NIGMS NIH HHS, National Science Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Alternative splicing (AS) of pre-mRNA plays a crucial role in tissue-specific gene regulation, with disease implications due to splicing defects. Predicting and manipulating AS can therefore uncover new regulatory mechanisms and aid in therapeutic design. We introduce TrASPr+BOS, a generative AI model with Bayesian Optimization for predicting and designing RNA for tissue-specific splicing outcomes. Transformer for Alternative Splicing Prediction (TrASPr) is a multi-transformer model that can handle different types of AS events and generalize to unseen cellular conditions. It then serves as an oracle, generating labeled data to train a Bayesian Optimization for Splicing (BOS) algorithm to design RNA for condition-specific splicing outcomes. We show TrASPr+BOS outperforms existing methods, enhancing tissue-specific AUPRC by up to 1.8-fold and capturing tissue-specific regulatory elements. We validate hundreds of predicted novel tissue-specific splicing variations and confirm new regulatory elements using dCas13. We envision TrASPr+BOS as a light yet accurate method researchers can probe or adopt for specific tasks.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

RNA SplicingAlternative SplicingComputational BiologyHumansBayes TheoremGenerative Artificial IntelligencePrediction AlgorithmsAlgorithmsRNA Precursors

Stichwörter der Autor:innen

Bayesian optimizationLSV-seqRNA splicingcomputational biologydeep generative modelshumansequence designsplicing codessystems biology

Substanzen

RNA Precursors

Autor:innen und Institutionen

  • Department of Computer and Information Science, School of Engineering, University of Pennsylvania, Philadelphia, United States.
  • Department of Genome Sciences, University of Washington, Seattle, United States.
  • Department of Genetics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, United States.
  • Division of Cancer Pathobiology, The Children's Hospital of Philadelphia, Philadelphia, United States.
  • Department of Pathology & Laboratory Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, United States.

ORCID-Profile: Di Wu, Yoseph Barash

Förderung

  • NLM NIH HHS
  • NIGMS NIH HHS
  • National Science Foundation
  • National Science Board
  • National Library of Medicine
  • National Institute of General Medical Sciences
  • CureBRCA
  • Basser Center for BRCA

Zitieren

APA

Wu, D., Maus, N., Jha, A., Yang, K., Wales-McGrath, B. D., Jewell, S., Tangiyan, A., Choi, P., Gardner, J. R., & Barash, Y. (2026). Generative modeling for RNA splicing prediction and design. eLife. https://doi.org/10.7554/elife.106043

Vancouver

Wu D, Maus N, Jha A, Yang K, Wales-McGrath BD, Jewell S, et al. Generative modeling for RNA splicing prediction and design. eLife. 2026. doi:10.7554/elife.106043

BibTeX
@article{Wu2026,
  author = {Wu, Di and Maus, Natalie and Jha, Anupama and Yang, Kevin and Wales-McGrath, Benjamin D and Jewell, San and Tangiyan, Anna and Choi, Peter and Gardner, Jake R and Barash, Yoseph},
  title = {Generative modeling for RNA splicing prediction and design},
  journal = {eLife},
  year = {2026},
  doi = {10.7554/elife.106043}
}
Kostenlos starten →