Last Updated on 2026-04-07
In the Kaggle competition Stanford RNA 3D Folding Part 2, Oda, a software engineer at our company, placed 10th and won a gold medal.
https://www.kaggle.com/competitions/stanford-rna-3d-folding-2/leaderboard
As in the previous competition [1], in addition to conventional sequence search methods [2], we applied Embedding-Based Alignment to enhance Template-Based Modeling.
We believe that another key factor behind this strong result was the improved stability achieved by incorporating several state-of-the-art models into our notebook, including RNAPro [2,3], OpenFold3 [4], and Boltz-2 [5].
In collaboration with its technology partner PEZY Computing, QuettaAI is strongly promoting the practical use of generative AI in healthcare fields such as drug discovery, drawing on its development experience in medical imaging diagnostic devices and genome analysis systems.
[1] https://www.quetta-ai.co.jp/?p=357&lang=en
[2] https://www.kaggle.com/code/jaejohn/rnapro-inference-with-tbm
[3] https://github.com/NVIDIA-Digital-Bio/RNAPro
[4] https://github.com/aqlaboratory/openfold-3
[5] https://github.com/jwohlwend/boltz