Deep learning models for RNA secondary structure prediction (probably) do not generalize across families
Open accessThe DOI matches its Crossref record; 8 of the 8 references it cites were found in a registry or an index.
Abstract
Motivation: The secondary structure of RNA is of importance to its function. Over the last few years, several papers attempted to use machine learning to improve de novo RNA secondary structure prediction. Many of these papers report impressive results for intra-family predictions but seldom address the much more difficult (and practical) inter-family problem. Results: We demonstrate that it is nearly trivial with convolutional neural networks to generate pseudo-free energy changes, modelled after structure mapping data that improve the accuracy of structure prediction for intra-family cases. We propose a more rigorous method for inter-family cross-validation that can be used to assess the performance of learning-based models. Using this method, we further demonstrate that intra-family performance is insufficient proof of generalization despite the widespread assumption in the literature and provide strong evidence that many existing learning-based models have not generalized inter-family. Availability and implementation: Source code and data are available at https://github.com/marcellszi/dl-rna. Supplementary information: Supplementary data are available at Bioinformatics online.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318
Learning representations by back-propagating errors
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986) · Nature · doi:10.1038/323533a0 · Cited by 31,892
Bidirectional recurrent neural networks
Schuster, M., & Paliwal, K. K. (1997) · IEEE Transactions on Signal Processing · doi:10.1109/78.650093 · Cited by 10,172
Lorenz, R., Bernhart, S. H., Höner zu Siederdissen, C., Tafer, H., Flamm, C., Stadler, P. F., & Hofacker, I. L. (2011) · Algorithms for Molecular Biology · doi:10.1186/1748-7188-6-26 · Cited by 5,098
Mathews, D. H., Sabina, J., Zuker, M., & Turner, D. H. (1999) · Journal of Molecular Biology · doi:10.1006/jmbi.1999.2700 · Cited by 3,795
Self-citationThe chemical repertoire of natural ribozymes
Doudna, J. A., & Cech, T. R. (2002) · Nature · doi:10.1038/418222a · Cited by 745
Rivas, E., Lang, R., & Eddy, S. R. (2012) · RNA · doi:10.1261/rna.030049.111 · Cited by 145
The four ingredients of single-sequence RNA secondary structure prediction. A unifying perspective
Rivas, E. (2013) · RNA Biology · doi:10.4161/rna.24971 · Cited by 57
Cited by
Multiple sequence alignment-based RNA language model and its application to structural inference
Zhang, Y., Lang, M., Jiang, J., Gao, Z., Xu, F., Litfin, T., Chen, K., Singh, J., Huang, X., Song, G., et al. (2024) · Nucleic Acids Research · doi:10.1093/nar/gkad1031 · Cited by 112
Szikszai, M., Magnus, M., Sanghi, S., Kadyan, S., Bouatta, N., & Rivas, E. (2024) · Journal of Molecular Biology · doi:10.1016/j.jmb.2024.168552 · Cited by 38
Self-citationDesigning molecular RNA switches with Restricted Boltzmann machines
Fernandez-de-Cossio-Diaz, J., Hardouin, P., Lyonnet du Moutier, F.-X., Di Gioacchino, A., Marchand, B., Ponty, Y., Sargueil, B., Monasson, R., & Cocco, S. (2025) · Nature Communications · doi:10.1038/s41467-025-66265-y · Cited by 20
BiRNA-BERT allows efficient RNA language modeling with adaptive tokenization
Tahmid, M. T., Shahgir, H. S., Mahbub, S., Dong, Y., & Bayzid, M. S. (2025) · Communications Biology · doi:10.1038/s42003-025-08982-0 · Cited by 16
sincFold: end-to-end learning of short- and long-range interactions in RNA secondary structure
(2024) · doi:10.1093/bib/bbae271 · Cited by 14
Metadata
- Authors
- Szikszai, M., Wise, M., Datta, A., Ward, M., & Mathews, D. H.
- Venue
- Bioinformatics
- Publisher
- Oxford University Press (OUP)
- Published
- 2022-06-24
- Volume
- 38
- Issue
- 16
- Pages
- 3892-3899
- ISSN
- 1367-4803 (ISSN-L)
- 1367-4811
- OpenAlex
- W4283447041
- Cited by
- 85
- References
- 8
- FWCI
- 4.76
- Type
- journal article
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.