Scholarlink

Background <p>Deep vein thrombosis (DVT) is a complex thrombotic disorder with multiple environmental and genetic determinants. Nevertheless, the intricacies of DVT regulatory mechanisms have thus far precluded comprehensive research on its multi-omics characteristics.</p> Methods <p>In this study, non-targeted metabolomics (<i>n</i> = 62) and transcriptomics (<i>n</i> = 17) were utilised to comprehensively analyse the peripheral blood changes of DVT patients and healthy subjects. The hub differentially expressed metabolites (DEMs) and differentially expressed genes (DEGs) involved in the pathophysiological process of DVT were screened based on two-way orthogonal partial least squares (O2PLS) and machine learning (ML). The predictive performance of transcriptomic features was evaluated by constructing a nomogram and validated using calibration curves, receiver operating characteristic (ROC) curves and validation sets (<i>n</i> = 18). Furthermore, single-cell RNA sequencing (scRNA-seq, <i>n</i> = 6) facilitates the revelation of intrinsic regulatory relationships between DEMs and DEGs through correlation analysis and inductively coupled plasma-mass spectrometry (ICP-MS).</p> Results <p>The metabolic and transcriptional profiles of DVT patients were found to be significantly different from those of healthy subjects, and a total of 85 DEMs and 193 DEGs were identified by difference analysis. Following an exploration of the relevant biological pathways via Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis, the top 25 DEMs and DEGs were identified by O2PLS. Following rigorous verification by the correlation network model and ML algorithm, it was determined that phosphatidylinositol (PI) and SLC39A11 exhibited significant characteristics. The nomogram demonstrated that SLC39A11 exhibited excellent prediction performance, with the area under the curve (AUC) value of 1.000. Moreover, SLC39A11 was found to be significantly up-regulated in the validation set (<i>p</i> &lt; 0.01). The SLC39A11 was localized to M1-like macrophages by scRNA-seq, and it was found that it was significantly related to PI metabolism in DVT by correlation analysis and ICP-MS (<i>p</i> &lt; 0.05).</p> Conclusions <p>This study demonstrated that the metabolic dysregulation of DVT was predominantly concentrated in lipid metabolism, as indicated by PI, and the gene dysregulation was represented by SLC39A11. SLC39A11 exerts an influence on the regulatory process of DVT through its interaction with zinc ion transport and PI metabolism.</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

Integrated multi-omics and machine learning identify an interaction between SLC39A11 and phosphoinositide metabolism in deep vein thrombosis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月22日

查看全文
  • Bao-Ze Pan,
  • Ming-Jun Jiang,
  • Jie Chen,
  • Dan Ning,
  • Jing Liang,
  • Zhi-He Deng,
  • Dong-Yang Luo,
  • Yang-Yi-Jing Wang,
  • Yao-Yang Zhong,
  • Xian-Peng Dai,
  • Li-Ming Deng,
  • Guo-Zuo Xiong,
  • Guo-Shan Bi

摘要

Background

Deep vein thrombosis (DVT) is a complex thrombotic disorder with multiple environmental and genetic determinants. Nevertheless, the intricacies of DVT regulatory mechanisms have thus far precluded comprehensive research on its multi-omics characteristics.

Methods

In this study, non-targeted metabolomics (n = 62) and transcriptomics (n = 17) were utilised to comprehensively analyse the peripheral blood changes of DVT patients and healthy subjects. The hub differentially expressed metabolites (DEMs) and differentially expressed genes (DEGs) involved in the pathophysiological process of DVT were screened based on two-way orthogonal partial least squares (O2PLS) and machine learning (ML). The predictive performance of transcriptomic features was evaluated by constructing a nomogram and validated using calibration curves, receiver operating characteristic (ROC) curves and validation sets (n = 18). Furthermore, single-cell RNA sequencing (scRNA-seq, n = 6) facilitates the revelation of intrinsic regulatory relationships between DEMs and DEGs through correlation analysis and inductively coupled plasma-mass spectrometry (ICP-MS).

Results

The metabolic and transcriptional profiles of DVT patients were found to be significantly different from those of healthy subjects, and a total of 85 DEMs and 193 DEGs were identified by difference analysis. Following an exploration of the relevant biological pathways via Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis, the top 25 DEMs and DEGs were identified by O2PLS. Following rigorous verification by the correlation network model and ML algorithm, it was determined that phosphatidylinositol (PI) and SLC39A11 exhibited significant characteristics. The nomogram demonstrated that SLC39A11 exhibited excellent prediction performance, with the area under the curve (AUC) value of 1.000. Moreover, SLC39A11 was found to be significantly up-regulated in the validation set (p < 0.01). The SLC39A11 was localized to M1-like macrophages by scRNA-seq, and it was found that it was significantly related to PI metabolism in DVT by correlation analysis and ICP-MS (p < 0.05).

Conclusions

This study demonstrated that the metabolic dysregulation of DVT was predominantly concentrated in lipid metabolism, as indicated by PI, and the gene dysregulation was represented by SLC39A11. SLC39A11 exerts an influence on the regulatory process of DVT through its interaction with zinc ion transport and PI metabolism.