Multi-omics integration and machine learning define an iron-sulfur cluster/zinc-binding protein prognostic signature in esophageal squamous cell carcinoma.

Authors

Wenhuan Chen, Tongtong Chen, Zhixuan Li, Wan Lin, Zefeng Xie, Hansheng Wu, Liyan Xu, Enmin Li, Hefeng Zhang, Yinwei Cheng

Published in

Mammalian genome : official journal of the International Mammalian Genome Society. Volume 37. Issue 1. Aug 24, 2026. Epub Aug 24, 2026.

Abstract

Esophageal squamous cell carcinoma (ESCC) is characterized by substantial intratumoral heterogeneity and poor clinical prognosis. Although metalloproteins are well-documented to drive ESCC malignant progression, incomplete functional annotation of this protein family significantly impedes the clinical translation of related research outcomes. This study reports the development and validation of a reliable prognostic model via integrating AlphaFold2-predicted iron-Sulfur (Fe-S) Cluster/Zinc (Zn)-binding proteins with ESCC multi-omics data. Nine differentially expressed AlphaFold2-predicted Fe-S/Zn-binding proteins significantly associated with ESCC prognosis were identified through integrated analysis of multi-omics and clinical data from public datasets and independent ESCC cohorts. After systematic evaluation of 117 machine learning combinations, a three-Fe-S/Zn-binding protein Prognostic Signature (FZPS) comprising YPEL5, MIB1 and ELAC2 was constructed, and validated as an independent predictor of poor overall survival across cohorts. High FZPS risk correlates with an immune-excluded, stress-adaptive phenotype with p21-driven inflammation and intrinsic immunotherapy resistance, while low-FZPS tumors harbor more actionable mutations and exhibit enhanced sensitivity to targeted therapy and immunotherapy. In vitro assays confirmed YPEL5 knockdown markedly suppresses ESCC cell viability, proliferation and migration. In conclusion, FZPS is a reliable independent prognostic biomarker guiding precision oncology practice for ESCC.

PMID:
42637882
Bibliographic data and abstract were imported from PubMed on 25 Aug 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.