Scholarlink

<p>Necrosis induced by sodium overload (NECSO) has recently been identified and characterized as a novel pattern of cellular necrosis that plays a protective role against energy depletion in cells or tissues. Nonetheless, the underlying mechanism of NECSO in lung adenocarcinoma (LUAD) remains to be elucidated. In this study, we collected 1652 NECSO-related genes (NECSORGs) from the GenesCards database. Subsequently, we integrated multi-omics data utilizing ten clustering algorithms to identify the consensus LUAD subtypes, which were subsequently employed three machine learning algorithms to develop a consensus necrosis induced by sodium overload-related signature (CNECSORS) for LUAD patients. Through multi-omics consensus clustering, we identified two LUAD subtypes, and observed that CS2 patients exhibited significantly lower overall survival than those in CS1 (<i>P</i> &lt; 0.001). The CNECSORS model comprises five characteristic genes (CPS1, KRT6A, CHIT1, NTS, and LYPD3). High-CNECSORS group demonstrated a substantially lower overall survival compared to low-CNECSORS group in the TCGA-LUAD cohort (<i>P</i> &lt; 0.001), GSE72094 cohort (<i>P</i> &lt; 0.001), GSE31210 cohort (<i>P</i> = 0.044), GSE68465 cohort (<i>P</i> = 0.001), GSE78220 cohort (<i>P</i> = 0.00017), GSE135222 cohort (<i>P</i> &lt; 0.0001), and GSE91061 cohort (<i>P</i> = 0.00024). The results of C-index and decision analysis showed that the nomograms delivered substantially higher clinical utility for LUAD patients than relying solely on the CNECSORS model. Excitingly, High-CNECSORS group exhibited greater responsiveness to chemotherapeutic drugs. Therefore, this study offers fresh perspectives on personalized prognosis prediction methods for LUAD patients by combining multi-omics data with advanced machine learning techniques, ultimately enhancing the precision of therapeutic strategies.</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 machine learning and multi-omics analysis develop a consensus necrosis induced by sodium overload-related signature for predicting prognosis and drug sensitivity for lung adenocarcinoma

  • 期刊论文

  • 发表日期：

    2025年11月20日

查看全文
  • Chunhong Li,
  • Wangshang Qin,
  • Yuhua Mao,
  • Jiahua Hu,
  • Shirong Nong,
  • Yi Liu

摘要

Necrosis induced by sodium overload (NECSO) has recently been identified and characterized as a novel pattern of cellular necrosis that plays a protective role against energy depletion in cells or tissues. Nonetheless, the underlying mechanism of NECSO in lung adenocarcinoma (LUAD) remains to be elucidated. In this study, we collected 1652 NECSO-related genes (NECSORGs) from the GenesCards database. Subsequently, we integrated multi-omics data utilizing ten clustering algorithms to identify the consensus LUAD subtypes, which were subsequently employed three machine learning algorithms to develop a consensus necrosis induced by sodium overload-related signature (CNECSORS) for LUAD patients. Through multi-omics consensus clustering, we identified two LUAD subtypes, and observed that CS2 patients exhibited significantly lower overall survival than those in CS1 (P < 0.001). The CNECSORS model comprises five characteristic genes (CPS1, KRT6A, CHIT1, NTS, and LYPD3). High-CNECSORS group demonstrated a substantially lower overall survival compared to low-CNECSORS group in the TCGA-LUAD cohort (P < 0.001), GSE72094 cohort (P < 0.001), GSE31210 cohort (P = 0.044), GSE68465 cohort (P = 0.001), GSE78220 cohort (P = 0.00017), GSE135222 cohort (P < 0.0001), and GSE91061 cohort (P = 0.00024). The results of C-index and decision analysis showed that the nomograms delivered substantially higher clinical utility for LUAD patients than relying solely on the CNECSORS model. Excitingly, High-CNECSORS group exhibited greater responsiveness to chemotherapeutic drugs. Therefore, this study offers fresh perspectives on personalized prognosis prediction methods for LUAD patients by combining multi-omics data with advanced machine learning techniques, ultimately enhancing the precision of therapeutic strategies.