Template-based RNA structure prediction advanced through a blind code competition

Automatically predicting RNA 3D structure from sequence remains an unsolved challenge in biology and biotechnology. Here, we describe a Kaggle code competition engaging over 1700 teams and 43 previously unreleased structures to tackle this challenge. The top three submitted algorithms achieved scores within statistical error of the winners of the recent CASP16 competition. Unexpectedly, the top Kaggle strategy involved a pipeline for discovering 3D templates, without the use of deep learning. We integrated this template-modeling pipeline and other Kaggle strategies to develop a single model RNAPro that retrospectively outperformed individual Kaggle models on the same test set. These results suggest a growing importance of template-based modeling in RNA structure prediction.

Authors

Institutions

Publication Details

Journal
bioRxiv (Cold Spring Harbor Laboratory)
Published
2025-12-30
DOI
https://doi.org/10.64898/2025.12.30.696949
Primary Topic
RNA and protein synthesis mechanisms
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Template-based RNA structure prediction advanced through a blind code competition

Emine Küçükbenli, Boris Rudolfs, Joseph J Piccirilli, Maggie Demkin et al.
bioRxiv (Cold Spring Harbor Laboratory)
RNA and protein synthesis mechanisms
article

Template-based RNA structure prediction advanced through a blind code competition

Emine Küçükbenli, Boris Rudolfs, Joseph J Piccirilli, Maggie Demkin, Danny Kowerko, Alejo Paullier, Toshiyuki Oda, Zhichao Miao, Cher Keng Heng, Youhan Lee, Hyunjin Kim, Andrew Favor, Navtej Toor, Jian Zou, Daniel B. Haack, Shanshan Li, Shi‐Jie Chen, Kaiming Zhang, Bin Zhao, Zhe Su, Zi Yang, E. Westhof, S. Zou, Rhiju Das, Shun Kuraishi, Andriy Kryshtafovych, Hoa Nguyen, Atsushi Yoshizawa, John Moult, Arunodhayan Sampathkumar, Haowei Li, Raul Enrique Gomez, Meng Chen, Tatsuya Hirano, Sicheng Zhang, Tsu-An Hsu, Shekhar Jadhav, Chaitanya K. Joshi, Kenji Ogawa, Rachael C. Kretsch, Jeanine G. Gezelle, Shujun He, Yibao Hong, Christian Munley, Yoshita Srivastava, Anna-Lena Steckelberg, G Gangadhareswar Rao, Alissa M. Hummer, Walter Reade, Anna Lewicka, Theo Viel, Huey-Long Chen, Raehyun Kim, Yehyun Kim, Zi Yang, Deepak Koirala
article en

Abstract

Automatically predicting RNA 3D structure from sequence remains an unsolved challenge in biology and biotechnology. Here, we describe a Kaggle code competition engaging over 1700 teams and 43 previously unreleased structures to tackle this challenge. The top three submitted algorithms achieved scores within statistical error of the winners of the recent CASP16 competition. Unexpectedly, the top Kaggle strategy involved a pipeline for discovering 3D templates, without the use of deep learning. We integrated this template-modeling pipeline and other Kaggle strategies to develop a single model RNAPro that retrospectively outperformed individual Kaggle models on the same test set. These results suggest a growing importance of template-based modeling in RNA structure prediction.

bioRxiv (Cold Spring Harbor Laboratory)
Uppsala University (SE), University of Science and Technology of China (CN), Google (United States) (US), Centre National de la Recherche Scientifique (FR), Howard Hughes Medical Institute (US), Institut Pasteur (FR), American International Group (United States) (US), University of Washington (US), NTT (Japan) (JP), University of Cambridge (GB), Sichuan University (CN), Chemnitz University of Technology (DE), Dalian University of Technology (CN), Science for Life Laboratory (SE), SLAC National Accelerator Laboratory (US), Italian Institute of Technology (IT), Yale University (US), University of California San Diego (US), University of Chicago (US), Guangzhou Experimental Station (CN), European Molecular Biology Laboratory (FR), Mohamed bin Zayed University of Artificial Intelligence (AE), Nvidia (United States) (US), Architecture et Réactivité de l'arN (FR), NewGen Therapeutics (United States) (US), Toyama University Hospital (JP), Institut de Biologie Moléculaire et Cellulaire (FR), Zhejiang University-University of Edinburgh Institute (CN), Detection Limit (United States) (US), Peptidream (Japan) (JP), University of Chinese Academy of Sciences (CN), Institute for Bioscience and Biotechnology Research, Université de Strasbourg (FR), Universidad de Montevideo (UY), University of Missouri (US), Columbia University (US), University of Maryland, Baltimore County (US), University of California, Davis (US), Texas A&M University (US), Stanford University (US)
National Science Foundation, Howard Hughes Medical Institute, U.S. Department of Energy, Nvidia, European Synchrotron Radiation Facility, National Natural Science Foundation of China, Technische Universität Dresden, National Institutes of Health, Office of Science, Argonne National Laboratory, SLAC National Accelerator Laboratory
Openalex Percentile: Top 35%
RNA and protein synthesis mechanisms
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.