EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Identification of diagnostic biomarkers for neonatal necrotizing enterocolitis via machine learning screening and single‐cell virtual gene knockout validation

Xue Liu, Wenqiang Sun, Zengxue Hu, Yihui Li, Jingtao Bian, Xueping Zhu

Journal of Cell Communication and Signaling · 2026-06 · 20 (2)

DOI: 10.1002/ccs3.70081 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Cell Communication and Signaling, online erschienen am 28.05.2026. Verfasst von Xue Liu und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Neonatology Children's Hospital of Soochow University Suzhou China. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Neonatal necrotizing enterocolitis (NEC) continues to be the most severe gastrointestinal emergency affecting preterm infants, with reported mortality rates ranging from 20% to 30%. The absence of distinct early biomarkers results in delayed intervention and unfavorable outcomes, and the molecular mechanisms underlying the dysregulated immune response in NEC remain incompletely defined. We integrated five gene expression omnibus datasets (36 NEC cases and 34 controls) using weighted gene co‐expression network analysis and systematically evaluated 113 machine learning algorithm combinations. The best‐performing model (RF + XGBoost) identified candidate diagnostic genes, which were validated through independent bulk RNA‐sequencing cohorts, single‐cell RNA sequencing (11,308 intestinal cells), single‐cell virtual gene knockout (scTenifoldKnk), and experimental models including mice and the intestinal epithelial cells cell line. Combinatorial in silico perturbation was further performed using Geneformer, and a ferroptosis gene panel was validated by qRT‐polymerase chain reaction in the mouse NEC model. The RF + XGBoost model achieved high diagnostic accuracy (area under the curve = 0.979, 95% CI: 0.940–1.000). Three key biomarkers were identified: SLC26A3, CCL20, and CXCL5. Multi‐platform validation showed consistent downregulation of SLC26A3 and upregulation of CCL20 and CXCL5 in NEC (all p < 0.001). Single‐cell analyses revealed cell‐type‐specific dysregulation, with CCL20 markedly elevated in macrophages (log2FC = 3.85) and CXCL5 broadly upregulated across enterocytes, macrophages, and fibroblasts. Immune profiling demonstrated elevated proportions of M0 macrophages and activated mast cells, alongside reduced naive B cells and naive CD4 T cells, and CXCL5 expression was strongly correlated with neutrophil infiltration ( r = 0.88, p < 0.001). Virtual knockout analysis revealed that perturbation of CCL20 and CXCL5 produced overlapping downstream networks converging on major histocompatibility complex class II–related antigen presentation genes, whereas SLC26A3 knockout perturbed a distinct set of epithelial barrier and innate immune genes. FTH1 was the sole gene perturbed across all three knockouts, implicating ferroptosis as a potential convergence point in NEC pathogenesis. Geneformer‐based combinatorial perturbation indicated subadditive rather than synergistic interactions among the three biomarkers, with SLC26A3 appearing as the dominant node. Extension to a 9‐gene ferroptosis panel showed coordinated directional shifts across the tested genes, and qPCR measurements in the NEC mouse ileum were consistent with the predicted expression changes for the five measured genes. SLC26A3, CCL20, and CXCL5 constitute a candidate molecular signature for early NEC diagnosis. Combinatorial perturbation analysis suggests subadditive rather than synergistic interactions among the three biomarkers, with SLC26A3 appearing as the dominant node. Virtual knockout network analyses suggest that CCL20 and CXCL5 share downstream regulatory circuits linked to antigen presentation, while SLC26A3 primarily perturbs the epithelial barrier and innate immune genes. Ferroptosis emerged as a coordinated multi‐node convergence point, with computational predictions corroborated by qPCR measurements in a mouse NEC model. These findings provide a framework for a mechanistic study and potential targeted intervention in NEC.

Schlagwörter

Stichwörter der Autor:innen

WGCNAdiagnostic biomarkersferroptosisimmune cell infiltrationmachine learningneonatal necrotizing enterocolitissingle‐cell RNA sequencingsingle‐cell virtual knockout

Autor:innen und Institutionen

  • Department of Neonatology Children's Hospital of Soochow University Suzhou China.
  • Suzhou Medical College Soochow University Suzhou China.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Liu, X., Sun, W., Hu, Z., Li, Y., Bian, J., & Zhu, X. (2026). Identification of diagnostic biomarkers for neonatal necrotizing enterocolitis via machine learning screening and single‐cell virtual gene knockout validation. Journal of Cell Communication and Signaling, 20(2). https://doi.org/10.1002/ccs3.70081

Vancouver

Liu X, Sun W, Hu Z, Li Y, Bian J, Zhu X. Identification of diagnostic biomarkers for neonatal necrotizing enterocolitis via machine learning screening and single‐cell virtual gene knockout validation. Journal of Cell Communication and Signaling. 2026;20(2). doi:10.1002/ccs3.70081

BibTeX
@article{Liu2026,
  author = {Liu, Xue and Sun, Wenqiang and Hu, Zengxue and Li, Yihui and Bian, Jingtao and Zhu, Xueping},
  title = {Identification of diagnostic biomarkers for neonatal necrotizing enterocolitis via machine learning screening and single‐cell virtual gene knockout validation},
  journal = {Journal of Cell Communication and Signaling},
  year = {2026},
  volume = {20},
  number = {2},
  doi = {10.1002/ccs3.70081}
}
Kostenlos starten →