Stanford-backed RNA folding challenge returns as AI tackles biology’s next hard problem

A second phase of a major Stanford-led Kaggle competition is now underway, pushing machine learning teams to predict RNA 3D structures using sequence data alone, as researchers target one of molecular biology’s most persistent challenges.

A new phase of the Stanford RNA 3D Folding Challenge is now live on Kaggle, inviting researchers and machine learning teams to develop models that can predict the three-dimensional structure of RNA molecules using only their sequences. The competition builds on a first phase that marked a milestone for the field and raises the technical bar with more complex targets and stricter evaluation.

The challenge sits at the intersection of AI research, life sciences, and skills development, as RNA structure prediction remains far less mature than protein folding despite its importance to understanding disease, treatment development, and fundamental biology.

RNA plays a central role in how cells function, but predicting how RNA folds into functional three-dimensional structures remains difficult. Unlike protein structure prediction, where AI systems have delivered major gains in recent years, RNA modeling has been limited by sparse data and the inherent complexity of RNA folding.

The first Stanford RNA 3D Folding Challenge demonstrated that fully automated machine learning models could match human experts for the first time. Part 2 builds on that result by introducing harder targets, including RNA molecules with no available structural templates, and by applying a revised evaluation framework designed to reward higher accuracy.

The competition is organized through a global collaboration involving experimental RNA structural biologists, Stanford University School of Medicine, and the AI@HHMI initiative of the Howard Hughes Medical Institute. It is timed to surface new approaches ahead of the seventeenth Critical Assessment of Structure Prediction, scheduled for April 2025.

Kaggle hosts research-grade competition and evaluation

The challenge is hosted on Kaggle, a platform widely used by the machine learning community for research competitions, benchmarking, and skills development. Kaggle provides the infrastructure for code-based submissions, leaderboards, and evaluation, positioning the competition as both a research exercise and a public testing ground for new modeling approaches.

Participants are required to submit notebooks that generate five predicted structures for each RNA sequence in the test set. Submissions are scored using TM-score, a standard structural similarity metric that compares predicted structures with experimentally determined reference structures. To discourage shortcuts, only residues that align correctly by numbering are rewarded, and scores are averaged across the best of five predictions for each target.

The organizers say this scoring approach is intended to push models toward precise structural accuracy rather than partial alignment or template reuse.

Stakes include prizes and scientific authorship

The competition runs on a fixed timeline, with entries open until March 18, 2026, and final submissions due on March 25. A private leaderboard will be released the following week after additional evaluation.

Prizes total seventy-five thousand dollars, with fifty thousand dollars awarded to the top-ranked team. In addition to cash awards, top-performing participants will be invited to contribute their code and model descriptions to a peer-reviewed scientific paper summarizing the competition’s outcomes, a significant incentive for academic and research-focused teams.

The return of the Stanford RNA 3D Folding Challenge highlights the role of open competitions in advancing difficult scientific problems while also developing applied AI skills. Unlike benchmark-only tasks, the challenge emphasizes end-to-end modeling, reproducibility, and alignment with experimental data.

ETIH Innovation Awards 2026

The ETIH Innovation Awards 2026 are now open and recognize education technology organizations delivering measurable impact across K–12, higher education, and lifelong learning. The awards are open to entries from the UK, the Americas, and internationally, with submissions assessed on evidence of outcomes and real-world application.