RNAmigos2: accelerated structure-based RNA virtual screening with deep graph learning

Abstract RNAs are a vast reservoir of untapped drug targets. Structure-based virtual screening (VS) identifies candidate molecules by leveraging binding site information, traditionally using molecular docking simulations. However, docking struggles to scale with large compound libraries and RNA targ...

Full description

Saved in:
Bibliographic Details
Main Authors: Juan G. Carvajal-Patiño, Vincent Mallet, David Becerra, Luis Fernando Niño Vasquez, Carlos Oliver, Jérôme Waldispühl
Format: Article
Language:English
Published: Nature Portfolio 2025-03-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-025-57852-0
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849389824824311808
author Juan G. Carvajal-Patiño
Vincent Mallet
David Becerra
Luis Fernando Niño Vasquez
Carlos Oliver
Jérôme Waldispühl
author_facet Juan G. Carvajal-Patiño
Vincent Mallet
David Becerra
Luis Fernando Niño Vasquez
Carlos Oliver
Jérôme Waldispühl
author_sort Juan G. Carvajal-Patiño
collection DOAJ
description Abstract RNAs are a vast reservoir of untapped drug targets. Structure-based virtual screening (VS) identifies candidate molecules by leveraging binding site information, traditionally using molecular docking simulations. However, docking struggles to scale with large compound libraries and RNA targets. Machine learning offers a solution but remains underdeveloped for RNA due to limited data and practical evaluations. We introduce a data-driven VS pipeline tailored for RNA, utilizing coarse-grained 3D modeling, synthetic data augmentation, and RNA-specific self-supervision. Our model achieves a 10,000x speedup over docking while ranking active compounds in the top 2.8% on structurally distinct test sets. It is robust to binding site variations and successfully screens unseen RNA riboswitches in a 20,000-compound in-vitro microarray, with a mean enrichment factor of 2.93 at 1%. This marks the first experimentally validated success of structure-based deep learning for RNA VS.
format Article
id doaj-art-d227bb65640b4d0b87b27b5e14f9dc1a
institution Kabale University
issn 2041-1723
language English
publishDate 2025-03-01
publisher Nature Portfolio
record_format Article
series Nature Communications
spelling doaj-art-d227bb65640b4d0b87b27b5e14f9dc1a2025-08-20T03:41:50ZengNature PortfolioNature Communications2041-17232025-03-0116111210.1038/s41467-025-57852-0RNAmigos2: accelerated structure-based RNA virtual screening with deep graph learningJuan G. Carvajal-Patiño0Vincent Mallet1David Becerra2Luis Fernando Niño Vasquez3Carlos Oliver4Jérôme Waldispühl5School of Computer Science, McGill UniversityLIX, Ecole PolytechniqueSchool of Computer Science, McGill UniversityUniversidad Nacional de Colombia - Sede Bogotá - Facultad de Ingeniería - Depto. de Ingeniería de Sistemas e IndustrialMax Planck Institute of BiochemistrySchool of Computer Science, McGill UniversityAbstract RNAs are a vast reservoir of untapped drug targets. Structure-based virtual screening (VS) identifies candidate molecules by leveraging binding site information, traditionally using molecular docking simulations. However, docking struggles to scale with large compound libraries and RNA targets. Machine learning offers a solution but remains underdeveloped for RNA due to limited data and practical evaluations. We introduce a data-driven VS pipeline tailored for RNA, utilizing coarse-grained 3D modeling, synthetic data augmentation, and RNA-specific self-supervision. Our model achieves a 10,000x speedup over docking while ranking active compounds in the top 2.8% on structurally distinct test sets. It is robust to binding site variations and successfully screens unseen RNA riboswitches in a 20,000-compound in-vitro microarray, with a mean enrichment factor of 2.93 at 1%. This marks the first experimentally validated success of structure-based deep learning for RNA VS.https://doi.org/10.1038/s41467-025-57852-0
spellingShingle Juan G. Carvajal-Patiño
Vincent Mallet
David Becerra
Luis Fernando Niño Vasquez
Carlos Oliver
Jérôme Waldispühl
RNAmigos2: accelerated structure-based RNA virtual screening with deep graph learning
Nature Communications
title RNAmigos2: accelerated structure-based RNA virtual screening with deep graph learning
title_full RNAmigos2: accelerated structure-based RNA virtual screening with deep graph learning
title_fullStr RNAmigos2: accelerated structure-based RNA virtual screening with deep graph learning
title_full_unstemmed RNAmigos2: accelerated structure-based RNA virtual screening with deep graph learning
title_short RNAmigos2: accelerated structure-based RNA virtual screening with deep graph learning
title_sort rnamigos2 accelerated structure based rna virtual screening with deep graph learning
url https://doi.org/10.1038/s41467-025-57852-0
work_keys_str_mv AT juangcarvajalpatino rnamigos2acceleratedstructurebasedrnavirtualscreeningwithdeepgraphlearning
AT vincentmallet rnamigos2acceleratedstructurebasedrnavirtualscreeningwithdeepgraphlearning
AT davidbecerra rnamigos2acceleratedstructurebasedrnavirtualscreeningwithdeepgraphlearning
AT luisfernandoninovasquez rnamigos2acceleratedstructurebasedrnavirtualscreeningwithdeepgraphlearning
AT carlosoliver rnamigos2acceleratedstructurebasedrnavirtualscreeningwithdeepgraphlearning
AT jeromewaldispuhl rnamigos2acceleratedstructurebasedrnavirtualscreeningwithdeepgraphlearning