Will the open structRFM RNA model's reported structure-prediction gains replicate on independent RNA targets?
structRFM research authors
This is a first-record evidence baseline. It does not establish independent verification of every research result or completion of future commitments.
Claim and evidence
A peer-reviewed Nature paper reports structRFM trained on about 21 million sequence–structure pairs and describes structure-aware representations for multiple RNA tasks. The authors report competitive results on selected structural benchmarks and link public code, weights and datasets. Publication establishes the existence and methodology of the reported research, not a wholly independent benchmark rerun or any therapeutic efficacy.
Limitations and uncertainty
Some structure labels are computationally inferred. The AlphaFold 3 comparisons are for defined subsets, not all RNA systems. Public weights and code do not imply that every upstream data component has an audited open-source license. No prospective experimental or clinical benefit was established.
Explicit acceptance criteria
- Reproduce selected RNA benchmark scores with fixed versions and documented test partitions.
- Probe novel RNA families and sequence lengths not represented in pretraining.
- Require independent wet-lab validation before any real-world biological outcome claim.
Dated baseline events
The entries below capture original source and publication milestones. They do not imply that later claims have already been fulfilled or independently replicated.