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

Youhan Lee; Shujun He; Toshiyuki Oda; G. John Rao; Yehyun Kim; Raehyun Kim; Hyunjin Kim; Cher Keng Heng; Danny Kowerko; Haowei Li; Hoa Nguyen; Arunodhayan Sampath Kumar; Raul Enrique Gomez; Meng Chen; Atsushi Yoshizawa; Shun Kuraishi; Kenji Ogawa; Shuxian Zou; Alejo Paullier; Bin Zhao; Huey-Long Chen; Tsu-An Hsu; Tatsuya Hirano; Wah Chiu; Jeanine Gezelle; Daniel B. Haack; Yibao Hong; Shekhar Jadhav; Deepak Koirala; Rachael C. Kretsch

doi:10.64898/2025.12.30.696949

BiophysicsbioRxivDec 30, 2025

Preprint

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

Youhan Lee iD, Shujun He iD, Toshiyuki Oda iD, G. John Rao, Yehyun Kim iD, Raehyun Kim iD, Hyunjin Kim iD, Cher Keng Heng, Danny Kowerko iD, Haowei Li, Hoa Nguyen iD, Arunodhayan Sampath Kumar iD, Raul Enrique Gomez, Meng Chen iD, Atsushi Yoshizawa iD, Shun Kuraishi iD, Kenji Ogawa iD, Shuxian Zou, Alejo Paullier, Bin Zhao iD, Huey-Long Chen, Tsu-An Hsu, Tatsuya Hirano, Wah Chiu iD, Jeanine Gezelle iD, , Rhiju Das iD

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.

PDF DOI