Investigating the impact of inflammatory response-related genes on renal fibrosis diagnosis: a machine learning-based study with experimental validation

Abstract

Renal fibrosis plays a crucial role in the progression of renal diseases, yet the lack of effective diagnostic markers poses challenges in scientific and clinical practices. In this study, we employed machine learning techniques to identify potential biomarkers for renal fibrosis. Utilizing two datasets from the GEO database, we applied LASSO, SVM-RFE and RF algorithms to screen for differentially expressed genes related to inflammatory responses between the renal fibrosis group and the control group. As a result, we identified four genes (CCL5, IFITM1, RIPK2, and TNFAIP6) as promising diagnostic indicators for renal fibrosis. These genes were further validated through in vivo experiments and immunohistochemistry, demonstrating their utility as reliable markers for assessing renal fibrosis. Additionally, we conducted a comprehensive analysis to explore the relationship between these candidate biomarkers, immunity, and drug sensitivity. Integrating these findings, we developed a nomogram with a high discriminative ability, achieving a concordance index of 0.933, enabling the prediction of disease risk in patients with renal fibrosis. Overall, our study presents a predictive model for renal fibrosis and highlights the significance of four potential biomarkers, facilitating clinical diagnosis and personalized treatment. This finding presents valuable insights for advancing precision medicine approaches in the management of renal fibrosis.

Communicated by Ramaswamy H. Sarma

Keywords:

• Renal fibrosis
• inflammatory response
• immune cell infiltration
• diagnostic markers
• machine learning

Author contributions

Conceptualization, Yangyang Guo; methodology, Ziwei Yuan, and Zujian Hu; software, Yang-yang Guo; validation, Yuanyuan Gao, and Penghua Yan; writ-ing—original draft preparation, Ziwei Yuan; writing—review and editing, Xuejia Yang; visualization, Fan Zheng; Investigation, Kai Hong, Kenan Cen, and Yifeng Mai; project administration, Yongheng Bai, Yangyang Guo and Jingzong Zhou; All authors have read and agreed to the published version of the manuscript.

Availability of data and materials

The datasets used for analysis in this study are publicly available online. The web address and original sources for each dataset are listed in the article material.

Competing interests

The authors affirm that they have no competing interests that would prevent this paper from being published.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This study was supported by the Ningbo University Institute of Geriatrics (LNBYJS-2021).