AIDO.RNA: Unlocking RNA’s Secrets with AI Foundation Models
Type | research |
---|---|
Area | AIDigital Twin |
Published(YearMonth) | 2411 |
Source | https://www.biorxiv.org/content/10.1101/2024.11.28.625345v1 |
Tag | newsletter |
Checkbox | |
Date(of entry) |
RNA, once considered a mere intermediary between DNA and proteins, now stands at the forefront of modern biology, critical for precision medicine, genetic engineering, and understanding cellular processes. To tackle RNA's complexity, researchers developed AIDO.RNA, a 1.6-billion-parameter AI model trained on 42 million non-coding RNA (ncRNA) sequences at single-nucleotide resolution. AIDO.RNA excels in tasks like RNA structure prediction, genetic regulation, molecular function analysis, and RNA sequence design, offering insights beyond the capabilities of protein-focused models. By modeling RNA’s unique role in protein translation, AIDO.RNA demonstrates the generality of biological sequence modeling and its potential to improve biomolecular understanding. This tool advances RNA research and accelerates innovations in drug design and biomolecular engineering. Models and code are publicly accessible, fostering collaborative progress in RNA science.