ORIGINAL RESEARCH article
Front. Immunol.
Sec. Inflammation
Volume 16 - 2025 | doi: 10.3389/fimmu.2025.1698740
This article is part of the Research TopicExploring Cardiovascular and Cerebrovascular Diseases Interaction with Inflammation: Biomarkers, Drug Targets, and Personalized Treatments through Multi-omics Data Integration, Volume IIView all 5 articles
Development of a Diagnostic Model for MASLD and Identification of Daidzein as the potential drug using bioinformatics analysis and experiments
Provisionally accepted
- 1Shuguang Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai, China
- 2Tongji University Dongfang Hospital, Shanghai, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Background Metabolic dysfunction-associated steatotic liver disease (MASLD) is now the predominant chronic liver disease globally, yet effective therapeutic strategies remain elusive. Methods MASLD-related datasets were download from GEO. Subsequently, genes associated with MASLD were found through the intersection of differentially expressed genes and WGCNA. Then, key candidate genes were further screened using 113 machine learning algorithms and their diagnostic value was evaluated using ROC curve analysis across multiple datasets. Genes are then screened by Shapley Additive exPlanations (SHAP) analysis. Molecular docking (MD) and molecular dynamics simulations (MDS) were employed to validate the interaction between Daidzein and Enolase 3 (ENO3). Finally, an in vitro fatty liver cell model was constructed to validate the "Enrichr" platform to identify poteitial drugs for MASLD. Results 62 MASLD-DEGs were finally identified. The optimal predictive model for MASLD was the 17-gene signature (IGFBP1, ENO3, SOCS2, GADD45G, NR4A2, RTP4, RAB26, CRYAA, PPP1R3C,MCAM, IL6, IER3, RTP3, NR4A1, CCL5, FOS, JUNB) selected through combined glmBoost+GBM algorithms, which was demonstrated robust predictive performance. SHAP analysis suggested that ENO3 may be the most prominent genes associated with MASLD severity. More importantly, we measured the effect of daidzein on improving lipid accumulation in vitro model. Conclusion We developed a predictive model for MASLD and identified ENO3 as a key predictive gene. Furthermore, we discovered that daidzein may serve as a potential therapeutic agent for MASLD. Through in vitro studies, we further confirmed that daidzein alleviates lipid deposition and improves MASLD by modulating the ENO3/PPAR signaling pathway.
Keywords: Metabolic dysfunction-associated steatotic liver disease, machine learning, Shap, ENO3, daidzein
Received: 04 Sep 2025; Accepted: 09 Oct 2025.
Copyright: © 2025 Wang, Zhang, Wang, Liu, Kong, Zhou and Qu. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Lihong Qu, 1905365@tongji.edu.cn
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.