Scholarlink

Study Design <p>Bioinformatics analysis.</p> Objectives <p>To explore the diagnostic signatures for patients with spinal cord injury (SCI).</p> Setting <p>Shenyang, China.</p> Methods <p>After DEGs screening, WGCNA was employed to screen the SCI-related genes. Then the SCI-related genes were intersected with DEGs and PANoptosis-related genes, and DPRGs were obtained, followed by PPI network construction. Diagnostic genes were identified by machine learning, followed by nomogram construction, immune infiltration analysis, GSEA, mRNA-miRNA and mRNA-TF networks establishment, and potential drugs screening.</p> Results <p>Total 277 PANoptosis-related genes,&#xa0;3269 DEGs, and 1417 SCI-related genes were identified, and intersection analysis yielded 28 intersection genes, which were considered DPRGs. Subsequently, seven diagnostic genes were identified by machine learning, namely FASLG, AVEN, PSMB3, EXOG, CASP4, CASP5, and TLR4. The CASP4 and TLR4 were both enriched in Toll-like receptor signaling pathway and NOD-like receptor signaling pathway, etc. Seventeen immune cells shown significant difference between SCI and control samples, including macrophage, activated dendritic cell, and neutrophil, etc. CASP4 and TLR4 were strongly associated with macrophage, activated dendritic cell and neutrophil, etc. The nomogram of CASP4 and TLR4 was constructed with powerful predictive accuracy. Finally, the potential drugs were predicted, including emricasan, neoceptin-3, and resatorvid, etc.</p> Conclusions <p>Seven PANoptosis-related genes, namely FASLG, AVEN, PSMB3, EXOG, CASP4, CASP5, and TLR4, and the constructed nomogram based on CASP4 and TLR4 might viable diagnostic signatures for SCI, and conducive to the prevention and diagnosis of SCI in clinical practice.</p> Sponsorship <p>This work was supported by Liaoning Provincial Natural Science Foundation of China (No. 2024-MSLH-538). I want to reiterate that there is no prior publication of figures or tables and no conflict of interest in the submission of this manuscript.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine learning-based identification of potential diagnostic signatures in spinal cord injury

  • 期刊论文

  • 发表日期：

    2026年2月27日

查看全文
  • Zheng Wang,
  • Haojun Sun,
  • Bin Liu,
  • Yu Wang,
  • Danni Wang,
  • Wenfeng Han

摘要

Study Design

Bioinformatics analysis.

Objectives

To explore the diagnostic signatures for patients with spinal cord injury (SCI).

Setting

Shenyang, China.

Methods

After DEGs screening, WGCNA was employed to screen the SCI-related genes. Then the SCI-related genes were intersected with DEGs and PANoptosis-related genes, and DPRGs were obtained, followed by PPI network construction. Diagnostic genes were identified by machine learning, followed by nomogram construction, immune infiltration analysis, GSEA, mRNA-miRNA and mRNA-TF networks establishment, and potential drugs screening.

Results

Total 277 PANoptosis-related genes, 3269 DEGs, and 1417 SCI-related genes were identified, and intersection analysis yielded 28 intersection genes, which were considered DPRGs. Subsequently, seven diagnostic genes were identified by machine learning, namely FASLG, AVEN, PSMB3, EXOG, CASP4, CASP5, and TLR4. The CASP4 and TLR4 were both enriched in Toll-like receptor signaling pathway and NOD-like receptor signaling pathway, etc. Seventeen immune cells shown significant difference between SCI and control samples, including macrophage, activated dendritic cell, and neutrophil, etc. CASP4 and TLR4 were strongly associated with macrophage, activated dendritic cell and neutrophil, etc. The nomogram of CASP4 and TLR4 was constructed with powerful predictive accuracy. Finally, the potential drugs were predicted, including emricasan, neoceptin-3, and resatorvid, etc.

Conclusions

Seven PANoptosis-related genes, namely FASLG, AVEN, PSMB3, EXOG, CASP4, CASP5, and TLR4, and the constructed nomogram based on CASP4 and TLR4 might viable diagnostic signatures for SCI, and conducive to the prevention and diagnosis of SCI in clinical practice.

Sponsorship

This work was supported by Liaoning Provincial Natural Science Foundation of China (No. 2024-MSLH-538). I want to reiterate that there is no prior publication of figures or tables and no conflict of interest in the submission of this manuscript.