Machine learning for RNA secondary structure prediction: a review of current methods and challenges
RNA · 2026-04 · 32 (4) · 443-456
DOI: 10.1261/rna.080840.125 ↗Kurz gefasst
Übersichtsarbeit in der Zeitschrift RNA, online erschienen am 23.01.2026. Verfasst von Giuseppe Sacco und 2 weiteren Autor:innen. Beteiligte Einrichtung: Scuola Internazionale Superiore di Studi Avanzati, SISSA, Trieste 34136, Italy. Hauptthemen laut MEDLINE: RNA, Nucleic Acid Conformation, Machine Learning, Computational Biology. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 94 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
Predicting the secondary structure of RNA is a core challenge in computational biology, essential for understanding molecular function and designing novel therapeutics. The field has evolved from foundational but accuracy-limited thermodynamic approaches to a new data-driven paradigm dominated by machine learning and deep learning. These models learn folding patterns directly from data, leading to significant performance gains. This review surveys the modern landscape of these methods, covering single-sequence, evolutionary-based, and hybrid models that blend machine learning with biophysics. A central theme is the field's “generalization crisis,” where powerful models were found to fail on new RNA families, prompting a community-wide shift to stricter, homology-aware benchmarking. In response to the underlying challenge of data scarcity, RNA foundation models have emerged, learning from massive, unlabeled sequence corpora to improve generalization. Finally, we look ahead to the next set of major hurdles—including the accurate prediction of complex motifs like pseudoknots, scaling to kilobase-length transcripts, incorporating the chemical diversity of modified nucleotides, and shifting the prediction target from static structures to the dynamic ensembles that better capture biological function. We also highlight the need for a standardized, prospective benchmarking system to ensure unbiased validation and accelerate progress.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Scuola Internazionale Superiore di Studi Avanzati, SISSA, Trieste 34136, Italy.
- Scuola Internazionale Superiore di Studi Avanzati, SISSA, Trieste 34136, Italy gsanguin@sissa.it.
ORCID-Profile: Giuseppe Sacco, Giovanni Bussi, Guido Sanguinetti
Zitieren
APA
Sacco, G., Bussi, G., & Sanguinetti, G. (2026). Machine learning for RNA secondary structure prediction: a review of current methods and challenges. RNA, 32(4), 443-456. https://doi.org/10.1261/rna.080840.125
Vancouver
Sacco G, Bussi G, Sanguinetti G. Machine learning for RNA secondary structure prediction: a review of current methods and challenges. RNA. 2026;32(4):443-456. doi:10.1261/rna.080840.125
BibTeX
@article{Sacco2026,
author = {Sacco, Giuseppe and Bussi, Giovanni and Sanguinetti, Guido},
title = {Machine learning for RNA secondary structure prediction: a review of current methods and challenges},
journal = {RNA},
year = {2026},
volume = {32},
number = {4},
pages = {443--456},
doi = {10.1261/rna.080840.125}
}
EUVIMED