Deep learning coupled with scalable domain-specific structural validation expands RNA virus discovery from metatranscriptomes.
6 affiliations
- 1Environmental Microbiome and Biotechnology Laboratory, Center of Synthetic Biology and Integrated Bioengineering, School of Engineering, Westlake University, Hangzhou, China.
- 2Research Center for Industries of the Future, Zhejiang Provincial Key Laboratory of Intelligent Low-Carbon Biosynthesis, Westlake University, Hangzhou, China.
- 3Center for Infectious Disease Research, Westlake Laboratory of Life Sciences and Biomedicine, School of Life Sciences, Westlake University, Hangzhou, China.
- 4Environmental Microbiome and Biotechnology Laboratory, Center of Synthetic Biology and Integrated Bioengineering, School of Engineering, Westlake University, Hangzhou, China. [email protected].
- 5Research Center for Industries of the Future, Zhejiang Provincial Key Laboratory of Intelligent Low-Carbon Biosynthesis, Westlake University, Hangzhou, China. [email protected].
- 6Center for Infectious Disease Research, Westlake Laboratory of Life Sciences and Biomedicine, School of Life Sciences, Westlake University, Hangzhou, China. [email protected].
Abstract
The discovery of RNA viruses from metatranscriptomic data remains challenging due to extreme sequence divergence and length heterogeneity, ranging from massive polyproteins to short assembly artifacts. We present Rider, a two-stage framework that synergizes deep representation learning with scalable domain-specific structural validation. In the first stage, Rider leverages a compact pretrained protein language model to broadly detect highly divergent viral signals. To overcome the inherent input length limitations of language models and accurately capture catalytic core domains, the second stage employs an overlapping sliding-window structure prediction strategy. These structures are then aligned against our curated database of non-redundant RdRp (RNA-dependent RNA polymerase) domains. This domain-centric design detects RdRp domains embedded within long polyproteins and truncated fragments. Across >10,000 metatranscriptomes spanning diverse ecosystems, Rider achieves sensitivity comparable to LucaProt, recalling >99% of previously cataloged RNA viruses. Rider identifies thousands of deeply divergent sequences and clades that elude existing methods. In a human inflammatory bowel disease (IBD) cohort, Rider agrees with other methods in detecting RNA viruses while extending detection to divergent lineages, providing a scalable pipeline for global RNA virome exploration.
Related
- El Agente Fármaco: A Scalable AI Scientist for Drug Discovery
- Structural diversity of human urinary N-glycans revealed by reversed-phase LC/MS/MS toward a resource for biomarker discovery
- Discovery of metabolite biomarkers for cerebral ischemia-reperfusion injury based on metabolomics: neurotoxic mechanism and functional validation of a flavopiridol -related metabolite
- Bone morphogenetic protein 4 meets artificial intelligence: a comprehensive review of AI-driven advances in drug discovery, structural modeling, and clinical translation
- AutoHMD: scalable and accurate validation of antibody-antigen complexes by heated molecular dynamics.
BibTeX
@article{luo2026deep,
title = {{Deep learning coupled with scalable domain-specific structural validation expands RNA virus discovery from metatranscriptomes.}},
author = {Gaoyang Luo and Zelin Zang and Ling Yuan and Jingbo Zhou and Ao Dong and Yufei Huang and Stan Z Li and Feng Ju},
journal = {Nature communications},
year = {2026},
month = sep,
doi = {10.1038/s41467-026-77183-y},
url = {https://europepmc.org/article/MED/42894013},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Deep learning coupled with scalable domain-specific structural validation expands RNA virus discovery from metatranscriptomes.
AU - Gaoyang Luo
AU - Zelin Zang
AU - Ling Yuan
AU - Jingbo Zhou
AU - Ao Dong
AU - Yufei Huang
AU - Stan Z Li
AU - Feng Ju
PY - 2026
DA - 2026/09/16
JO - Nature communications
DO - 10.1038/s41467-026-77183-y
UR - https://europepmc.org/article/MED/42894013
AB - The discovery of RNA viruses from metatranscriptomic data remains challenging due to extreme sequence divergence and length heterogeneity, ranging from massive polyproteins to short assembly artifacts. We present Rider, a two-stage framework that synergizes deep representation learning with scalable domain-specific structural validation. In the first stage, Rider leverages a compact pretrained protein language model to broadly detect highly divergent viral signals. To overcome the inherent input length limitations of language models and accurately capture catalytic core domains, the second stage employs an overlapping sliding-window structure prediction strategy. These structures are then aligned against our curated database of non-redundant RdRp (RNA-dependent RNA polymerase) domains. This domain-centric design detects RdRp domains embedded within long polyproteins and truncated fragments. Across >10,000 metatranscriptomes spanning diverse ecosystems, Rider achieves sensitivity comparable to LucaProt, recalling >99% of previously cataloged RNA viruses. Rider identifies thousands of deeply divergent sequences and clades that elude existing methods. In a human inflammatory bowel disease (IBD) cohort, Rider agrees with other methods in detecting RNA viruses while extending detection to divergent lineages, providing a scalable pipeline for global RNA virome exploration.
SN - 2041-1723
ER -