Deep learning on protein language model embeddings unlocks accurate prediction of protein solubility
Frontiers in Microbiology · 24.03.2026 · 17
DOI: 10.3389/fmicb.2026.1716930 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Microbiology, online erschienen am 24.03.2026. Verfasst von Jing Cui und 8 weiteren Autor:innen. Beteiligte Einrichtung: Key Laboratory of Specialty Agri-products Quality and Hazard Controlling Technology of Zhejiang Province, College of Life Sciences, China Jiliang Uni…. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Zhejiang Province, Basic Research Program of Jiangsu Province.
Abstract (Originalsprache)
Introduction
Protein misfolding is a major limitation in prokaryotic expression systems, which lack post-translational modifications and exhibit distinct intracellular environments. This severely hinders the functional expression of many heterologous proteins, especially in Escherichia coli . Accurate prediction of protein solubility is crucial for synthetic biology and protein engineering but remains a challenging task.
Methods
Here, we present DeepSolNet, a deep learning model that leverages advanced protein language models to enhance solubility prediction. DeepSolNet adopts a multi-module architecture, integrating contextual embeddings from ESM Cambrian with bidirectional long short-term memory networks, convolutional neural networks, and attention mechanisms.
Results
On the validation set, DeepSolNet achieved an accuracy of 0.75 and a Matthews correlation coefficient of 0.50 for soluble/insoluble protein classification. On an independently constructed test set containing gammabody, transglutaminase, and aldehyde dehydrogenase sequences, the model maintained high performance with an accuracy of 0.53, achieving state-of-the-art performance. Visualization analyses further showed that DeepSolNet is sensitive to key residues influencing protein solubility.
Discussion
These results demonstrate that DeepSolNet serves as a powerful and generalizable tool for large-scale protein design and expression optimization. The tool is freely available at https://github.com/wangxinglong1990/DeepSolNet .
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Key Laboratory of Specialty Agri-products Quality and Hazard Controlling Technology of Zhejiang Province, College of Life Sciences, China Jiliang University, Hangzhou, Zhejiang, China.
- School of Laboratory Medicine, Jilin Medical University, Jilin City, Jilin, China.
- Zhejiang Gongzheng Testing Center Co., Ltd, Hangzhou, Zhejiang, China.
- Medical Enzyme Engineering Center, CAS Key Lab of Bio-Medical Diagnostics, Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, China.
Förderung
- National Natural Science Foundation of China
- Natural Science Foundation of Zhejiang Province
- Basic Research Program of Jiangsu Province
Zitieren
APA
Cui, J., Jiang, Y., Wang, J., Jiang, H., Fang, J., Jiang, J., Chen, Q., Zhong, S., & Wang, X. (2026). Deep learning on protein language model embeddings unlocks accurate prediction of protein solubility. Frontiers in Microbiology, 17. https://doi.org/10.3389/fmicb.2026.1716930
Vancouver
Cui J, Jiang Y, Wang J, Jiang H, Fang J, Jiang J, et al. Deep learning on protein language model embeddings unlocks accurate prediction of protein solubility. Frontiers in Microbiology. 2026;17. doi:10.3389/fmicb.2026.1716930
BibTeX
@article{Cui2026,
author = {Cui, Jing and Jiang, Yong and Wang, Jing and Jiang, Han and Fang, Jiehong and Jiang, Jiankang and Chen, Qi and Zhong, Shihuan and Wang, Xinglong},
title = {Deep learning on protein language model embeddings unlocks accurate prediction of protein solubility},
journal = {Frontiers in Microbiology},
year = {2026},
volume = {17},
doi = {10.3389/fmicb.2026.1716930}
}
EUVIMED