[go: up one dir, main page]
Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

A 5′ UTR language model for decoding untranslated regions of mRNA and function predictions

Nature Machine Intelligence volume 6, pages 449–460 (2024)Cite this article

Subjects

An Author Correction to this article was published on 02 August 2024

This article has been updated

A preprint version of the article is available at bioRxiv.

Abstract

The 5′ untranslated region (UTR), a regulatory region at the beginning of a messenger RNA (mRNA) molecule, plays a crucial role in regulating the translation process and affects the protein expression level. Language models have showcased their effectiveness in decoding the functions of protein and genome sequences. Here, we introduce a language model for 5′ UTR, which we refer to as the UTR-LM. The UTR-LM is pretrained on endogenous 5′ UTRs from multiple species and is further augmented with supervised information including secondary structure and minimum free energy. We fine-tuned the UTR-LM in a variety of downstream tasks. The model outperformed the best known benchmark by up to 5% for predicting the mean ribosome loading, and by up to 8% for predicting the translation efficiency and the mRNA expression level. The model was also applied to identifying unannotated internal ribosome entry sites within the untranslated region and improved the area under the precision–recall curve from 0.37 to 0.52 compared to the best baseline. Further, we designed a library of 211 new 5′ UTRs with high predicted values of translation efficiency and evaluated them via a wet-laboratory assay. Experiment results confirmed that our top designs achieved a 32.5% increase in protein production level relative to well-established 5′ UTRs optimized for therapeutics.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on SpringerLink
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Overview of the UTR-LM model for 5′ UTR function prediction and design.
Fig. 2: Prediction of MRL on the basis of 5′ UTR sequences.
Fig. 3: Prediction of mRNA TE and EL for endogenous datasets.
Fig. 4: UTR-LM accurately identifies unannotated IRES.
Fig. 5: Experimental validation of UTR-LM model and top sequence designs generated by UTR-LM.
Fig. 6: The UTR language model recognizes underlying patterns and reveals known motif patterns.

Similar content being viewed by others

Data availability

The datasets are available and can be downloaded at https://codeocean.com/capsule/6711822 (ref. 39). This link includes training data for the pretrained model as well as datasets for various downstream tasks. Detailed statistics for these datasets are provided in Supplementary Discussion A. Source data are provided with this paper.

Code availability

The code is freely available at https://github.com/a96123155/UTR-LM (ref. 40) under the GNU General Public Licence Version 3 and the implemented demo can be found at https://codeocean.com/capsule/4214075/tree/v1 (ref. 39).

Change history

References

  1. Araujo, P. R. et al. Before it gets started: regulating translation at the 5′ UTR. Comp. Funct. Genomics 2012, 475731 (2012).

    Article  Google Scholar 

  2. Miao, Z., Tidu, A., Eriani, G. & Martin, F. Secondary structure of the SARS-CoV-2 5′-UTR. RNA Biol. 18, 447–456 (2021).

    Article  Google Scholar 

  3. Li, X., Kazan, H., Lipshitz, H. D. & Morris, Q. D. Finding the target sites of RNA-binding proteins. Wiley Interdiscip. Rev. RNA 5, 111–130 (2014).

    Article  Google Scholar 

  4. Zeraati, M. et al. Cancer-associated noncoding mutations affect RNA G-quadruplex-mediated regulation of gene expression. Sci. Rep. 7, 708 (2017).

    Article  Google Scholar 

  5. Karollus, A., Avsec, Ž. & Gagneur, J. Predicting mean ribosome load for 5′ UTR of any length using deep learning. PLoS Comput. Biol. 17, e1008982 (2021).

    Article  Google Scholar 

  6. Sample, P. J. et al. Human 5′ UTR design and variant effect prediction from a massively parallel translation assay. Nat. Biotechnol. 37, 803–809 (2019).

    Article  Google Scholar 

  7. Cao, J. et al. High-throughput 5′ UTR engineering for enhanced protein production in non-viral gene therapies. Nat. Commun. 12, 4138 (2021).

    Article  Google Scholar 

  8. Barazandeh, S., Ozden, F., Hincer, A., Seker, U. O. S. & Cicek, A. E. UTRGAN: learning to generate 5′ UTR sequences for optimized translation efficiency and gene expression. Preprint at bioRxiv https://doi.org/10.1101/2023.01.30.526198 (2023).

  9. Zheng, W. et al. Discovery of regulatory motifs in 5′ untranslated regions using interpretable multi-task learning models. Cell Syst. 14, 1103–1112.e6 (2023).

    Article  Google Scholar 

  10. Chen, J. et al. Interpretable RNA foundation model from unannotated data for highly accurate RNA structure and function predictions. Preprint at https://doi.org/10.48550/arxiv.2204.00300 (2022).

  11. Ozden, F., Barazandeh, S., Akboga, D., Seker, U. O. S. & Cicek, A. E. RNAGEN: a generative adversarial network-based model to generate synthetic RNA sequences to target proteins. Preprint at bioRxiv https://doi.org/10.1101/2023.07.11.548246 (2023).

  12. Akiyama, M. & Sakakibara, Y. Informative RNA base embedding for RNA structural alignment and clustering by deep representation learning. NAR Genom. Bioinform. 4, lqac012 (2022).

    Article  Google Scholar 

  13. Wang, J. & Gribskov, M. IRESpy: an XGBoost model for prediction of internal ribosome entry sites. BMC Bioinf. 20, 409 (2019).

    Article  Google Scholar 

  14. Kolekar, P., Pataskar, A., Kulkarni-Kale, U., Pal, J. & Kulkarni, A. IRESPred: web server for prediction of cellular and viral internal ribosome entry site (IRES). Sci. Rep. 6, 27436 (2016).

    Article  Google Scholar 

  15. Zhao, J. et al. IRESfinder: identifying RNA internal ribosome entry site in eukaryotic cell using framed k-mer features. J. Genet. Genomics 45, 403–406 (2018).

    Article  Google Scholar 

  16. Zhou, Y. et al. DeepCIP: a multimodal deep learning method for the prediction of internal ribosome entry sites of circRNAs. Comput. Biol. Med. 164, 107288 (2023).

    Article  Google Scholar 

  17. Zeng, C. et al. Leveraging mRNA sequences and nanoparticles to deliver SARS-CoV-2 antigens in vivo. Adv. Mater. 32, e2004452 (2020).

    Article  Google Scholar 

  18. Babendure, J. R., Babendure, J. L., Ding, J.-H. & Tsien, R. Y. Control of mammalian translation by mRNA structure near caps. RNA 12, 851–861 (2006).

    Article  Google Scholar 

  19. Hinnebusch, A. G., Ivanov, I. P. & Sonenberg, N. Translational control by 5′-untranslated regions of eukaryotic mRNAs. Science 352, 1413–1416 (2016).

    Article  Google Scholar 

  20. Calvo, S. E., Pagliarini, D. J. & Mootha, V. K. Upstream open reading frames cause widespread reduction of protein expression and are polymorphic among humans. Proc. Natl Acad. Sci. USA 106, 7507–7512 (2009).

    Article  Google Scholar 

  21. Zuccotti, P. & Modelska, A. Studying the translatome with polysome profiling. Post-Transcriptional Gene Regulation (ed Dassi, E.) 59–69 (2016).

  22. Whiffin, N. et al. Characterising the loss-of-function impact of 5′ untranslated region variants in 15,708 individuals. Nat. Commun. 11, 2523 (2020).

  23. Kozak, M. An analysis of 5′-noncoding sequences from 699 vertebrate messenger RNAs. Nucleic Acids Res. 15, 8125–8148 (1987).

    Article  Google Scholar 

  24. Kozak, M. Downstream secondary structure facilitates recognition of initiator codons by eukaryotic ribosomes. Proc. Natl Acad. Sci. USA 87, 8301–8305 (1990).

    Article  Google Scholar 

  25. Stoneley, M. & Willis, A. E. Cellular internal ribosome entry segments: structures, trans-acting factors and regulation of gene expression. Oncogene 23, 3200–3207 (2004).

    Article  Google Scholar 

  26. Weingarten-Gabbay, S. et al. Comparative genetics. Systematic discovery of cap-independent translation sequences in human and viral genomes. Science 351, aad4939 (2016).

    Article  Google Scholar 

  27. Zhao, J. et al. IRESbase: a comprehensive database of experimentally validated internal ribosome entry sites. Genom. Proteom. Bioinform. 18, 129–139 (2020).

    Article  Google Scholar 

  28. Mokrejs, M. et al. IRESite–a tool for the examination of viral and cellular internal ribosome entry sites. Nucleic Acids Res. 38, D131–D136 (2010).

    Article  Google Scholar 

  29. Kalvari, I. et al. Rfam 14: expanded coverage of metagenomic, viral and microRNA families. Nucleic Acids Res. 49, D192–D200 (2021).

    Article  Google Scholar 

  30. Leppek, K. et al. Combinatorial optimization of mRNA structure, stability, and translation for RNA-based therapeutics. Nat. Commun. 13, 1536 (2022).

    Article  Google Scholar 

  31. Gleason, A. C., Ghadge, G., Chen, J., Sonobe, Y. & Roos, R. P. Machine learning predicts translation initiation sites in neurologic diseases with nucleotide repeat expansions. PLoS ONE 17, e0256411 (2022).

    Article  Google Scholar 

  32. Hernández, G., Osnaya, V. G. & Pérez-Martínez, X. Conservation and variability of the AUG initiation codon context in eukaryotes. Trends Biochem. Sci. 44, 1009–1021 (2019).

    Article  Google Scholar 

  33. Cunningham, F. et al. Ensembl 2022. Nucleic Acids Res. 50, D988–D995 (2022).

    Article  Google Scholar 

  34. Vaswani, A. et al. Attention is all you need. In 31st Conference on Neural Information Processing Systems (NIPS, 2017).

  35. Sinha, K. et al. Masked language modeling and the distributional hypothesis: order word matters pre-training for little. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing 2888–2913 (2021).

  36. Lorenz, R. et al. ViennaRNA package 2.0. Algorithms Mol. Biol. 6, 26 (2011).

    Article  Google Scholar 

  37. Leppek, K., Das, R. & Barna, M. Functional 5′ UTR mRNA structures in eukaryotic translation regulation and how to find them. Nat. Rev. Mol. Cell Biol. 19, 158–174 (2018).

    Article  Google Scholar 

  38. Rao, R. M., Meier, J., Sercu, T., Ovchinnikov, S. & Rives, A. Transformer protein language models are unsupervised structure learners. International Conference on Learning Representations (ICLR, 2020).

  39. Chu, Y. et al. A 5′ UTR language model for decoding untranslated regions of mRNA and function predictions. Zenodo https://doi.org/10.5281/zenodo.10621605 (2024).

  40. Chu, Y. et al. UTR-LM GitHub https://github.com/a96123155/UTR-LM (2024).

Download references

Acknowledgements

This paper is partially supported by National Science Foundation grant no. 1953686 and partially supported by RVAC Medicines.

Author information

Author notes

  1. These authors contributed equally: Yanyi Chu, Dan Yu.

Authors and Affiliations

  1. Department of Electrical and Computer Engineering, Princeton University, Princeton, NJ, USA

    Yanyi Chu, Kaixuan Huang & Mengdi Wang

  2. Department of Pathology, School of Medicine, Stanford University, Stanford, CA, USA

    Yanyi Chu & Le Cong

  3. RVAC Medicines, Waltham, MA, USA

    Dan Yu & Yue Shen

  4. Zipcode Bio, Weston, MA, USA

    Yupeng Li & Jason Zhang

Authors
  1. Yanyi Chu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Dan Yu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yupeng Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Kaixuan Huang
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Yue Shen
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Le Cong
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Jason Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Mengdi Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Y.C. developed the UTR-LM model. D.Y. performed experimental validation. Y.L. produced in-house data. K.H. reviewed both the code and manuscript. Y.S. developed the web server. L.C. contributed to manuscript preparation. J.Z. initiated the experimental part of the project. M.W. led the entire project. All authors contributed to manuscript preparation.

Corresponding author

Correspondence to Mengdi Wang.

Ethics declarations

Competing interests

RVAC Medicines has submitted patent applications related to the designed UTR sequences. D.Y., Y.L. and Y.S. are affiliated with RVAC Medicines. J.Z. is affiliated with Zipcode Bio. Other authors have declared no conflicts of interest.

Peer review

Peer review information

Nature Machine Intelligence thanks Joshua W. K. Ho, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1

The computation flow of identifying patterns in 5′ UTR sequences based on attention scores.

Supplementary information

Supplementary Information

Supplementary Discussion, Figs. 1–10 and Tables 1–6.

Source data

Source Data Fig. 2

Statistical source data.

Source Data Fig. 3

Statistical source data.

Source Data Fig. 4

Statistical source data.

Source Data Fig. 5

Statistical source data.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Chu, Y., Yu, D., Li, Y. et al. A 5′ UTR language model for decoding untranslated regions of mRNA and function predictions. Nat Mach Intell 6, 449–460 (2024). https://doi.org/10.1038/s42256-024-00823-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s42256-024-00823-9

This article is cited by

Search

Advanced search

Quick links

Nature Briefing AI and Robotics

Sign up for the Nature Briefing: AI and Robotics newsletter — what matters in AI and robotics research, free to your inbox weekly.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing: AI and Robotics