Generative Modeling for RNA Splicing Predictions and Design

Overview

Journal bioRxiv

Date 2025 Feb 3

PMID 39896553

Authors

Di Wu

Natalie Maus

Anupama Jha

Kevin Yang

Benjamin D Wales-McGrath

San Jewell

Anna Tangiyan

Peter Choi

Jacob R Gardner

Yoseph Barash

Affiliations

Soon will be listed here.

Abstract

Alternative splicing (AS) of pre-mRNA plays a crucial role in tissue-specific gene regulation, with disease implications due to splicing defects. Predicting and manipulating AS can therefore uncover new regulatory mechanisms and aid in therapeutics design. We introduce TrASPr+BOS, a generative AI model with Bayesian Optimization for predicting and designing RNA for tissue-specific splicing outcomes. TrASPr is a multi-transformer model that can handle different types of AS events and generalize to unseen cellular conditions. It then serves as an oracle, generating labeled data to train a Bayesian Optimization for Splicing (BOS) algorithm to design RNA for condition-specific splicing outcomes. We show TrASPr+BOS outperforms existing methods, enhancing tissue-specific AUPRC by up to 2.4 fold and capturing tissue-specific regulatory elements. We validate hundreds of predicted novel tissue-specific splicing variations and confirm new regulatory elements using dCas13. We envision TrASPr+BOS as a light yet accurate method researchers can probe or adopt for specific tasks.

References

Winter R, Montanari F, Steffen A, Briem H, Noe F, Clevert D . Efficient multi-objective molecular optimization in a continuous latent space. Chem Sci. 2019; 10(34):8016-8024. PMC: 6836962. DOI: 10.1039/c9sc01928f. View

Husedzinovic A, Neumann B, Reymann J, Draeger-Meurer S, Chari A, Erfle H . The catalytically inactive tyrosine phosphatase HD-PTP/PTPN23 is a novel regulator of SMN complex localization. Mol Biol Cell. 2014; 26(2):161-71. PMC: 4294665. DOI: 10.1091/mbc.E14-06-1151. View

Bend R, Cohen L, Carter M, Lyons M, Niyazov D, Mikati M . Phenotype and mutation expansion of the PTPN23 associated disorder characterized by neurodevelopmental delay and structural brain abnormalities. Eur J Hum Genet. 2019; 28(1):76-87. PMC: 6906308. DOI: 10.1038/s41431-019-0487-1. View

Ugolino J, Fang S, Kubisch C, Monteiro M . Mutant Atp13a2 proteins involved in parkinsonism are degraded by ER-associated degradation and sensitize cells to ER-stress induced cell death. Hum Mol Genet. 2011; 20(18):3565-77. PMC: 3159557. DOI: 10.1093/hmg/ddr274. View

Ji Y, Zhou Z, Liu H, Davuluri R . DNABERT: pre-trained Bidirectional Encoder Representations from Transformers model for DNA-language in genome. Bioinformatics. 2021; 37(15):2112-2120. PMC: 11025658. DOI: 10.1093/bioinformatics/btab083. View

Zeng T, Li Y . Predicting RNA splicing from DNA sequence using Pangolin. Genome Biol. 2022; 23(1):103. PMC: 9022248. DOI: 10.1186/s13059-022-02664-4. View

Xu C, Zhang G, Wang X, Huang X, Zhang J, Han S . Ptpn23 Controls Cardiac T-Tubule Patterning by Promoting the Assembly of Dystrophin-Glycoprotein Complex. Circulation. 2024; 149(17):1375-1390. PMC: 11039371. DOI: 10.1161/CIRCULATIONAHA.123.065767. View

Maurin M, Ranjouri M, Megino-Luque C, Newberg J, Du D, Martin K . RBFOX2 deregulation promotes pancreatic cancer progression and metastasis through alternative splicing. Nat Commun. 2023; 14(1):8444. PMC: 10730836. DOI: 10.1038/s41467-023-44126-w. View

Wang Z, Burge C . Splicing regulation: from a parts list of regulatory elements to an integrated splicing code. RNA. 2008; 14(5):802-13. PMC: 2327353. DOI: 10.1261/rna.876308. View

10.

Avsec Z, Agarwal V, Visentin D, Ledsam J, Grabska-Barwinska A, Taylor K . Effective gene expression prediction from sequence by integrating long-range interactions. Nat Methods. 2021; 18(10):1196-1203. PMC: 8490152. DOI: 10.1038/s41592-021-01252-x. View

11.

Van Nostrand E, Freese P, Pratt G, Wang X, Wei X, Xiao R . A large-scale binding and functional map of human RNA-binding proteins. Nature. 2020; 583(7818):711-719. PMC: 7410833. DOI: 10.1038/s41586-020-2077-3. View

12.

Cheng J, Celik M, Kundaje A, Gagneur J . MTSplice predicts effects of genetic variants on tissue-specific splicing. Genome Biol. 2021; 22(1):94. PMC: 8011109. DOI: 10.1186/s13059-021-02273-7. View

13.

Griffiths R, Hernandez-Lobato J . Constrained Bayesian optimization for automatic chemical design using variational autoencoders. Chem Sci. 2020; 11(2):577-586. PMC: 7067240. DOI: 10.1039/c9sc04026a. View

14.

Singh S, Yeat N, Wang Y, Lin S, Kuo I, Wu K . PTPN23 ubiquitination by WDR4 suppresses EGFR and c-MET degradation to define a lung cancer therapeutic target. Cell Death Dis. 2023; 14(10):671. PMC: 10567730. DOI: 10.1038/s41419-023-06201-4. View

15.

Pan Q, Shai O, Lee L, Frey B, Blencowe B . Deep surveying of alternative splicing complexity in the human transcriptome by high-throughput sequencing. Nat Genet. 2008; 40(12):1413-5. DOI: 10.1038/ng.259. View

16.

Barash Y, Vaquero-Garcia J, Gonzalez-Vallinas J, Xiong H, Gao W, Lee L . AVISPA: a web tool for the prediction and analysis of alternative splicing. Genome Biol. 2013; 14(10):R114. PMC: 4014802. DOI: 10.1186/gb-2013-14-10-r114. View

17.

Yeo G, Burge C . Maximum entropy modeling of short sequence motifs with applications to RNA splicing signals. J Comput Biol. 2004; 11(2-3):377-94. DOI: 10.1089/1066527041410418. View

18.

Yang K, Islas N, Jewell S, Wu D, Jha A, Radens C . Machine learning-optimized targeted detection of alternative splicing. Nucleic Acids Res. 2024; 53(3). PMC: 11797022. DOI: 10.1093/nar/gkae1260. View

19.

Siepel A, Bejerano G, Pedersen J, Hinrichs A, Hou M, Rosenbloom K . Evolutionarily conserved elements in vertebrate, insect, worm, and yeast genomes. Genome Res. 2005; 15(8):1034-50. PMC: 1182216. DOI: 10.1101/gr.3715005. View

20.

Xiong H, Barash Y, Frey B . Bayesian prediction of tissue-regulated splicing using RNA sequence and cellular context. Bioinformatics. 2011; 27(18):2554-62. DOI: 10.1093/bioinformatics/btr444. View