RNA language models predict mutations that improve RNA function

Yekaterina Shulgina; Marena I Trinidad; Conner J Langeberg; Hunter Nisonoff; Seyone Chithrananda; Petr Skopintsev; Amos J Nissley; Jaymin Patel; Ron S Boger; Honglue Shi; Peter H Yoon; Erin E Doherty; Tara Pande; Aditya M Iyer; Jennifer A Doudna; Jamie H D Cate

doi:10.1038/s41467-024-54812-y

RNA language models predict mutations that improve RNA function

Nat Commun. 2024 Dec 5;15(1):10627. doi: 10.1038/s41467-024-54812-y.

Authors

Yekaterina Shulgina^#^{1

2

3}, Marena I Trinidad^#^{1

4}, Conner J Langeberg^#^{1

2

3}, Hunter Nisonoff^#⁵, Seyone Chithrananda^#^{1

6}, Petr Skopintsev^#^{1

3}, Amos J Nissley^#⁷, Jaymin Patel¹, Ron S Boger^{1

8}, Honglue Shi^{1

4}, Peter H Yoon^{1

2}, Erin E Doherty^{1

3}, Tara Pande⁶, Aditya M Iyer⁹, Jennifer A Doudna^{1

2

3

4

7

10

11}, Jamie H D Cate^#^{12

13

14

15

16}

Affiliations

¹ Innovative Genomics Institute, University of California, Berkeley, CA, USA.
² Department of Molecular and Cell Biology, University of California, Berkeley, CA, USA.
³ California Institute for Quantitative Biosciences, University of California, Berkeley, CA, USA.
⁴ Howard Hughes Medical Institute, University of California, Berkeley, CA, USA.
⁵ Center for Computational Biology, University of California, Berkeley, CA, USA.
⁶ Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, CA, USA.
⁷ Department of Chemistry, University of California, Berkeley, CA, USA.
⁸ Biophysics Graduate Program, University of California, Berkeley, CA, USA.
⁹ Department of Physics, University of California, Berkeley, CA, USA.
¹⁰ MBIB Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA.
¹¹ Gladstone Institutes, University of California, San Francisco, CA, USA.
¹² Innovative Genomics Institute, University of California, Berkeley, CA, USA. j-h-doudna-cate@berkeley.edu.
¹³ Department of Molecular and Cell Biology, University of California, Berkeley, CA, USA. j-h-doudna-cate@berkeley.edu.
¹⁴ California Institute for Quantitative Biosciences, University of California, Berkeley, CA, USA. j-h-doudna-cate@berkeley.edu.
¹⁵ Department of Chemistry, University of California, Berkeley, CA, USA. j-h-doudna-cate@berkeley.edu.
¹⁶ MBIB Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA. j-h-doudna-cate@berkeley.edu.

^# Contributed equally.

Abstract

Structured RNA lies at the heart of many central biological processes, from gene expression to catalysis. RNA structure prediction is not yet possible due to a lack of high-quality reference data associated with organismal phenotypes that could inform RNA function. We present GARNET (Gtdb Acquired RNa with Environmental Temperatures), a new database for RNA structural and functional analysis anchored to the Genome Taxonomy Database (GTDB). GARNET links RNA sequences to experimental and predicted optimal growth temperatures of GTDB reference organisms. Using GARNET, we develop sequence- and structure-aware RNA generative models, with overlapping triplet tokenization providing optimal encoding for a GPT-like model. Leveraging hyperthermophilic RNAs in GARNET and these RNA generative models, we identify mutations in ribosomal RNA that confer increased thermostability to the Escherichia coli ribosome. The GTDB-derived data and deep learning models presented here provide a foundation for understanding the connections between RNA sequence, structure, and function.

MeSH terms

Deep Learning
Escherichia coli* / genetics
Escherichia coli* / metabolism
Mutation*
Nucleic Acid Conformation*
RNA* / genetics
RNA* / metabolism
RNA, Ribosomal / chemistry
RNA, Ribosomal / genetics
Ribosomes / genetics
Ribosomes / metabolism

Substances

RNA
RNA, Ribosomal

Abstract

MeSH terms

Substances

Grants and funding