Scholarlink

<p>The global incidence of colorectal cancer (CRC) continues to rise, presenting a substantial disease burden worldwide, and most CRC patients are diagnosed at advanced stages and experience poor outcomes, whereas early-stage CRC patients have significantly better prognoses following radical treatment. In this study, we utilized a computational framework to integrate multi-omics data from CRC patients using the latest 10 different clustering algorithms, which were then employed 101 combinations derived from 10 different machine learning algorithms to develop a consensus machine learning-related signature (CMLRS). Using multi-omics consensus clustering, we distinguished four cancer subtypes (CSs) of CRC, and found that CS4 patients demonstrated the most favorable clinical outcomes. The developed CMLRS model comprises 11 prognosis-related differentially expressed genes (PRDEGs). This model demonstrated the capability to predict survival outcomes in the TCGA-COAD, GSE17536, GSE29621, GSE38832 and multiple immunotherapy cohorts, exhibiting high performance. Additionally, the CMLRS model showed substantial correlations with immune profiles, immunotherapy response, and chemotherapy. The findings were promising, as we noted that the low-CMLRS patients demonstrated heightened responsiveness to conventional chemotherapeutic drugs compared to the high-CMLRS patients, indicating that the CMLRS model may serve as a valuable predictor for determining chemotherapy strategies in CRC patients. Therefore, in-depth examination of data from multi-omics data can provide valuable insights and contribute to the refinement of the molecular classification of CRC. In addition, the CMLRS model acts as a crucial auxiliary tool in enhancing clinical decision-making and tailoring treatment strategies by offering important information about the molecular features of CRC patients.</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 refine molecular subtypes and clinical outcome for colorectal cancer

  • 期刊论文

  • 发表日期：

    2026年2月17日

查看全文
  • Chunhong Li,
  • Yuhua Mao,
  • Yi Liu,
  • Xiaoping Pang,
  • Jiahua Hu,
  • Minglin Ou

摘要

The global incidence of colorectal cancer (CRC) continues to rise, presenting a substantial disease burden worldwide, and most CRC patients are diagnosed at advanced stages and experience poor outcomes, whereas early-stage CRC patients have significantly better prognoses following radical treatment. In this study, we utilized a computational framework to integrate multi-omics data from CRC patients using the latest 10 different clustering algorithms, which were then employed 101 combinations derived from 10 different machine learning algorithms to develop a consensus machine learning-related signature (CMLRS). Using multi-omics consensus clustering, we distinguished four cancer subtypes (CSs) of CRC, and found that CS4 patients demonstrated the most favorable clinical outcomes. The developed CMLRS model comprises 11 prognosis-related differentially expressed genes (PRDEGs). This model demonstrated the capability to predict survival outcomes in the TCGA-COAD, GSE17536, GSE29621, GSE38832 and multiple immunotherapy cohorts, exhibiting high performance. Additionally, the CMLRS model showed substantial correlations with immune profiles, immunotherapy response, and chemotherapy. The findings were promising, as we noted that the low-CMLRS patients demonstrated heightened responsiveness to conventional chemotherapeutic drugs compared to the high-CMLRS patients, indicating that the CMLRS model may serve as a valuable predictor for determining chemotherapy strategies in CRC patients. Therefore, in-depth examination of data from multi-omics data can provide valuable insights and contribute to the refinement of the molecular classification of CRC. In addition, the CMLRS model acts as a crucial auxiliary tool in enhancing clinical decision-making and tailoring treatment strategies by offering important information about the molecular features of CRC patients.