MOTIVATION: Lipid nanoparticles (LNPs) are the most widely used vehicles for mRNA vaccine delivery. The structure of the lipids composing the LNPs can have a major impact on the effectiveness of the mRNA payload. Several properties should be optimized to improve delivery and expression including biodegradability, synthetic accessibility, and transfection efficiency. RESULTS: To optimize LNPs, we developed and tested models that enable the virtual screening of LNPs with high transfection efficiency. Our best method uses the lipid Simplified Molecular-Input Line-Entry System (SMILES) as inputs to a large language model. Large language model-generated embeddings are then used by a downstream gradient-boosting classifier. As we show, our method can more accurately predict lipid properties, which could lead to higher efficiency and reduced experimental time and costs. AVAILABILITY AND IMPLEMENTATION: Code and data links available at: https://github.com/Sanofi-Public/LipoBART.

DataSentics Brno 602 00 Czech Republic

Digital R and D Sanofi Cambridge MA 02141 United States

Digital R and D Sanofi Toronto ON M5V 1V6 Canada

mRNA Center of Excellence Marcy L'Etoile Sanofi 69280 France

mRNA Center of Excellence Sanofi Waltham MA 02451 United States

Zobrazit více v PubMed

Aimo L, Liechti R, Hyka-Nouspikel N  et al.  The SwissLipids knowledgebase for lipid biology. Bioinformatics  2015;31:2860–6. PubMed PMC

Baden LR, El Sahly HM, Essink B  et al.  Efficacy and safety of the mRNA-1273 SARS-CoV-2 vaccine. N Engl J Med  2021;384:403–16. PubMed PMC

Bjerrum E, Edwards L  MegaMolBart: Generally applicable chemical AI models with large-scale pretrained transformers. https://resources.nvidia.com/en-us-drug-discovery/gtcfall21-a31106 (7 December 2023, date last accessed).

Curtis A, Henderson M, Eygeris Y  et al.  Quantifying lipid nanoparticle-mediated GFP expression in the murine retina. Invest Ophthalmol Vis Sci  2023;64:5027.

Dara T, Vatanara A, Meybodi MN  et al.  Erythropoietin-loaded solid lipid nanoparticles: preparation, optimization, and in vivo evaluation. Colloids Surf B Biointerfaces  2019;178:307–16. PubMed

Ding DY, Zhang Y, Jia Y  et al.  Machine learning-guided lipid nanoparticle design for mRNA delivery. arXiv, arXiv: 2308.01402, 2023, preprint: not peer reviewed.

Duvenaud DK, Maclaurin D, Iparraguirre J  et al.  Convolutional networks on graphs for learning molecular fingerprints. Adv Neural Inf Process Syst  2015;28:2224–32.

Eygeris Y, Gupta M, Kim J  et al.  Chemistry of lipid nanoparticles for RNA delivery. Acc Chem Res  2021;55:2–12. PubMed

Gaulton A, Bellis LJ, Bento AP  et al.  ChEMBL: a large-scale bioactivity database for drug discovery. Nucleic Acids Res  2012;40:D1100–7. PubMed PMC

Gilmer J, Schoenholz SS, Riley PF  et al. Neural message passing for quantum chemistry. In: Precup D, Teh, YW. (eds) International Conference on Machine Learning, Sydney, Australia, 2017. 1263-72. Proceedings of Machine Learning Research.

Goh GB, Siegel C, Vishnu A  et al.  Chemception: a deep neural network with minimal chemistry knowledge matches the performance of expert-developed QSAR/QSPR models. arXiv, arXiv: 1706.06689, 2017, preprint: not peer reviewed.

Hajj KA, Whitehead KA.  Tools for translation: non-viral materials for therapeutic mRNA delivery. Nat Rev Mater  2017;2:1–17.

Han X, Zhang H, Butowska K  et al.  An ionizable lipid toolbox for RNA delivery. Nat Commun  2021;12:7233. PubMed PMC

Hinton GE, Srivastava N, Krizhevsky A  et al.  Improving neural networks by preventing co-adaptation of feature detectors. arXiv, arXiv: 1207.0580, 2012, preprint: not peer reviewed.

Hou X, Zaks T, Langer R  et al.  Lipid nanoparticles for mRNA delivery. Nat Rev Mater  2021;6:1078–94. PubMed PMC

Irwin R, Dimitriadis S, He J  et al.  Chemformer: a pre-trained transformer for computational chemistry. Mach Learn Sci Technol  2022;3:015022.

Kauffman KJ, Dorkin JR, Yang JH  et al.  Optimization of lipid nanoparticle formulations for mRNA delivery in vivo with fractional factorial and definitive screening designs. Nano Lett  2015;15:7300–6. 10.1021/acs.nanolett.5b02497 PubMed DOI

Kim J, Eygeris Y, Gupta M  et al.  Self-assembled mRNA vaccines. Adv Drug Deliv Rev  2021;170:83–112. PubMed PMC

Kon E, Elia U, Peer D.  Principles for designing an optimal mRNA lipid nanoparticle vaccine. Curr Opin Biotechnol  2022;73:329–36. 10.1016/j.copbio.2021.09.016. PubMed DOI PMC

Lewis M, Liu Y, Goyal N  et al.  Bart: denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension. arXiv, arXiv: 1910.13461, 2019, preprint: not peer reviewed.

Li B, Luo X, Deng B  et al.  An orthogonal array optimization of lipid-like nanoparticles for mRNA delivery in vivo. Nano Lett  2015;15:8099–107. PubMed PMC

Li S, Moayedpour S, Li R  et al.  Codonbert: large language models for mRNA design and optimization. bioRxiv, 10.1101/2023.09.09.556981, 2023. DOI

Li Y, Li R, Chakraborty A  et al.  Combinatorial library of cyclic benzylidene acetal-containing pH-responsive lipidoid nanoparticles for intracellular mRNA delivery. Bioconjug Chem  2020;31:1835–43. PubMed PMC

Lin Z, Akin H, Rao R  et al.  Language models of protein sequences at the scale of evolution enable accurate structure prediction. bioRxiv, 2022.

Liu S, Cheng Q, Wei T  et al.  Membrane-destabilizing ionizable phospholipids for organ-selective mRNA delivery and CRISPR–Cas gene editing. Nat Mater  2021;20:701–10. PubMed PMC

Meng C, Chen Z, Li G  et al.  Nanoplatforms for mRNA therapeutics. Adv Therap  2021;4:2000099.

Mouchlis VD, Afantitis A, Serra A  et al.  Advances in de novo drug design: from conventional to machine learning methods. Int J Mol Sci  2021;22:1676. PubMed PMC

Rajan K, Zielesny A, Steinbeck C.  DECIMER: towards deep learning for chemical image recognition. J Cheminform  2020;12:65–9. PubMed PMC

Rogers D, Hahn M.  Extended-connectivity fingerprints. J Chem Inf Model  2010;50:742–54. PubMed

Rong Y, Bian Y, Xu T  et al.  Self-supervised graph transformer on large-scale molecular data. Adv Neural Inf Process Syst  2020;33:12559–71.

Ross J, Belgodere B, Chenthamarakshan V  et al.  Large-scale chemical language representations capture molecular structure and properties. Nat Mach Intell  2022;4:1256–64.

Sterling T, Irwin JJ.  Zinc 15–ligand discovery for everyone. J Chem Inf Model  2015;55:2324–37. PubMed PMC

Sun D, Lu Z-R.  Structure and function of cationic and ionizable lipids for nucleic acid delivery. Pharm Res  2023;40:27–46. PubMed PMC

Ucak UV, Ashyrmamatov I, Lee J.  Improving the quality of chemical language model outcomes with atom-in-smiles tokenization. J Cheminform  2023;15:55. PubMed PMC

Winter B, Winter C, Schilling J  et al.  A smile is all you need: predicting limiting activity coefficients from smiles with natural language processing. Digit Discov  2022;1:859–69. PubMed PMC

Zhang X, Zhao W, Nguyen GN  et al.  Functionalized lipid-like nanoparticles for in vivo mRNA delivery and base editing. Sci Adv  2020;6(34):eabc2315. PubMed PMC

Computational Methods for Modeling Lipid-Mediated Active Pharmaceutical Ingredient Delivery