Open Access

Combination of machine learning and protein‑protein interaction network established one ATM‑DPP4‑TXN ferroptotic diagnostic model with experimental validation

  • Authors:
    • Mengze Wu
    • Zhao Zou
    • Yuce Peng
    • Suxin Luo
  • View Affiliations

  • Published online on: July 2, 2025     https://doi.org/10.3892/mmr.2025.13604
  • Article Number: 239
  • Copyright: © Wu et al. This is an open access article distributed under the terms of Creative Commons Attribution License.

Metrics: Total Views: 0 (Spandidos Publications: | PMC Statistics: )
Total PDF Downloads: 0 (Spandidos Publications: | PMC Statistics: )


Abstract

Ferroptosis and lethal sepsis are interlinked, although this association remains largely unknown to clinical panels. Sepsis is characterized by dysfunction of the inflammatory microenvironment. Most septic biomarkers lack independent validation, and a comprehensive diagnosis comprising biomarker assessment combined with clinical evaluation may improve sepsis management. Targeting ferroptosis regulators may offer new hope for uncovering the inflammatory machinery and for developing novel diagnostic methods for sepsis, and bioinformatics analyses are a valuable tool to investigate this further. In the present study, septic datasets were obtained from the Gene Expression Omnibus database. Differentially expressed genes (DEGs) were subsequently introduced in enrichment analyses and intersected with ferroptotic genes for acquiring ferroptosis‑related DEGs (FRDEGs). A protein‑protein interaction network (PPIN) was then constructed to retain hub‑FRDEGs, and this was imported into three machine learning algorithms. A nomogram based on the logistic regression model was subsequently built and validated in silico. CIBERSORT and single‑sample gene set enrichment analysis were used to carry out an analysis of the immune microenvironment, and inflammatory associations with the hub‑FRDEGs were examined. A cellular model was subsequently applied to substantiate the results of the bioinformatic analyses. A total of 94 FRDEGs were obtained from the overlap of 4,410 DEGs and 506 ferroptotic genes. One PPIN of FRDEGs was constructed to identify 38 hub‑FRDEGs, and the three machine learning algorithms were subsequently analyzed, which resulted in the identification of three hub‑FRDEGs, namely ataxia telangiectasia mutated, dipeptidyl peptidase 4 and thioredoxin. One diagnostic nomogram was advanced and scrutinized for its diagnostic accuracy. The functions and pathways of the DEGs were revealed to be mainly concentrated on the immune response and cellular transportation. A notably wide discrepancy was demonstrated to exist between the hub‑FRDEGs and the immunocytes. In conclusion, three potential hub‑FRDEGs connected with sepsis were identified in the present study. Their diagnostic accuracy and immune association demonstrated that ferroptosis is implicated in the inflammatory dysfunction of sepsis, and based on these findings, novel strategies for pharmacological interference and improving diagnostic utility may be developed to facilitate improved management of sepsis.
View Figures
View References

Related Articles

Journal Cover

September-2025
Volume 32 Issue 3

Print ISSN: 1791-2997
Online ISSN:1791-3004

Sign up for eToc alerts

Recommend to Library

Copy and paste a formatted citation
x
Spandidos Publications style
Wu M, Zou Z, Peng Y and Luo S: Combination of machine learning and protein‑protein interaction network established one ATM‑DPP4‑TXN ferroptotic diagnostic model with experimental validation. Mol Med Rep 32: 239, 2025.
APA
Wu, M., Zou, Z., Peng, Y., & Luo, S. (2025). Combination of machine learning and protein‑protein interaction network established one ATM‑DPP4‑TXN ferroptotic diagnostic model with experimental validation. Molecular Medicine Reports, 32, 239. https://doi.org/10.3892/mmr.2025.13604
MLA
Wu, M., Zou, Z., Peng, Y., Luo, S."Combination of machine learning and protein‑protein interaction network established one ATM‑DPP4‑TXN ferroptotic diagnostic model with experimental validation". Molecular Medicine Reports 32.3 (2025): 239.
Chicago
Wu, M., Zou, Z., Peng, Y., Luo, S."Combination of machine learning and protein‑protein interaction network established one ATM‑DPP4‑TXN ferroptotic diagnostic model with experimental validation". Molecular Medicine Reports 32, no. 3 (2025): 239. https://doi.org/10.3892/mmr.2025.13604