Interpretable AI-guided discovery of degradable ionizable lipids with enhanced hepatic mRNA delivery in vivo

Abstract

Lipid nanoparticles (LNPs) enable RNA medicines, yet ionizable-lipid discovery remains largely empirical, particularly when in vitro activity must be translated into functional hepatic delivery in vivo. Here we establish an interpretable discovery framework that combines modular degradable lipid chemistry, high-throughput phenotyping, machine learning, virtual library expansion, and independent in vivo validation to define a learnable ionizable-lipid design space. Starting from a 520-member library, we train an XGBoost model on firefly luciferase (Fluc) expression in HeLa cells and use physicochemical descriptors and molecular fingerprints to prospectively prioritize candidates with higher predicted in vitro transfection activity across a 10,920-member virtual space. In vivo phenotyping subsequently identifies PA97MD101 as a lead ionizable lipid. In mice, PA97MD101 enhances hepatic mRNA expression and Cre-dependent recombination and supports Cas9/sgTtr-mediated TTR suppression. Its performance is associated with pH-dependent membrane activity and reduced lysosomal colocalization. These findings support further evaluation of PA97MD101 for functional hepatic RNA delivery.

Plug-and-play assembly of biodegradable ionizable lipids for potent mRNA delivery and gene editing in vivo
Xuexiang Han, Ying Xu, Adele S. Ricciardi +12
Proceedings of the National Academy of Sciences26 May 2026Peer-reviewedCC BY-NC-ND15 citedMolecular biology & gene regulationOpenAlex
STRIVE: interpretable multimodal subtype discovery in Women’s cancers using an attention-weighted joint variational autoencoder
Zunaira Mehmood, Divya Sharma
Frontiers in bioinformatics16 Oct 2026Peer-reviewedCC BYFull textMolecular biology & gene regulationEurope PMC
CPRISP: an interpretable deep learning framework predicting circRNA-RBP interactions and guiding therapeutic discovery in neurodegeneration
Yongtian Wang, Jiahao Li, Xinmeng Liu +2
Nature Communications28 Oct 2026Peer-reviewedCC BY 4.0Molecular biology & gene regulationCrossref
Engineering morphology of tetrasulfide-incorporated mesoporous silica nanoparticles for enhanced mRNA delivery
Z. Cao, S. Saarela, J. Wang +7
bioRxiv3 Nov 2026PreprintCC BYFull textGeneral biochemistrybioRxiv
Discovery of Putative Phosphorylation Sites in Cre1 and Ace1 as Engineering Targets for Enhanced Cellulase Production in Trichoderma reesei
Jiahui Liu, Sunxing Wang, Xin Gao +2
International Journal of Molecular Sciences15 Oct 2026Peer-reviewedCC BY 4.0Full textMolecular biology & gene regulationCrossrefEurope PMC

Cite

BibTeX RIS

BibTeX

@article{zhang2026interpretable,
  title = {{Interpretable AI-guided discovery of degradable ionizable lipids with enhanced hepatic mRNA delivery in vivo}},
  author = {Zhen Zhang and Yongbo Ren and Deqiang Kong and Jindong Feng and Kaiyi Hu and Xinyun Xu and Fukang Deng and Peixia Li and Jiuao Wang and Wenhui Huang and Guodan Wei and Yang Su and Ying Tan and Shuixing Zhang and Xiaodan Sun and Feiyu Kang and Hao Yan},
  journal = {Nature Communications},
  year = {2026},
  month = oct,
  doi = {10.1038/s41467-026-78635-1},
  url = {https://www.nature.com/articles/s41467-026-78635-1},
  note = {CC BY 4.0},
}

RIS

TY  - JOUR
TI  - Interpretable AI-guided discovery of degradable ionizable lipids with enhanced hepatic mRNA delivery in vivo
AU  - Zhen Zhang
AU  - Yongbo Ren
AU  - Deqiang Kong
AU  - Jindong Feng
AU  - Kaiyi Hu
AU  - Xinyun Xu
AU  - Fukang Deng
AU  - Peixia Li
AU  - Jiuao Wang
AU  - Wenhui Huang
AU  - Guodan Wei
AU  - Yang Su
AU  - Ying Tan
AU  - Shuixing Zhang
AU  - Xiaodan Sun
AU  - Feiyu Kang
AU  - Hao Yan
PY  - 2026
DA  - 2026/10/12
JO  - Nature Communications
DO  - 10.1038/s41467-026-78635-1
UR  - https://www.nature.com/articles/s41467-026-78635-1
AB  - Lipid nanoparticles (LNPs) enable RNA medicines, yet ionizable-lipid discovery remains largely empirical, particularly when in vitro activity must be translated into functional hepatic delivery in vivo. Here we establish an interpretable discovery framework that combines modular degradable lipid chemistry, high-throughput phenotyping, machine learning, virtual library expansion, and independent in vivo validation to define a learnable ionizable-lipid design space. Starting from a 520-member library, we train an XGBoost model on firefly luciferase (Fluc) expression in HeLa cells and use physicochemical descriptors and molecular fingerprints to prospectively prioritize candidates with higher predicted in vitro transfection activity across a 10,920-member virtual space. In vivo phenotyping subsequently identifies PA97MD101 as a lead ionizable lipid. In mice, PA97MD101 enhances hepatic mRNA expression and Cre-dependent recombination and supports Cas9/sgTtr-mediated TTR suppression. Its performance is associated with pH-dependent membrane activity and reduced lysosomal colocalization. These findings support further evaluation of PA97MD101 for functional hepatic RNA delivery.
SN  - 2041-1723
ER  -

Abstract

Related

Cite