Scholarlink

Biological phenomena are based on the precise and accurate cooperation of a non-random combination of molecules implicated in several pathways and networks. In the view of precision medicine, the plethora of omics data accrued sheds light on the comprehension of molecules cooperation. However, these data bring noise and redundancy that it is necessary to consider during the data analysis. A combination of omics data resources, integrated to parameterize mechanistic models, and multiphase Ensemble Feature Selection (EFS) is proposed. Through EFS, we characterized the metabolic heterogeneity of three distinct glycolysis-associated clusters (GACs) in colorectal cancer. Our study reveals that the EFS-derived genetic signatures associated with each GAC group also characterize three glycolysis profiles previously identified. GAC1 demonstrated unique separation, while GAC2 and GAC3 exhibited overlapping characteristics.

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

Synergy Between Mechanistic Modelling and Ensemble Feature Selection Approaches to Explore Multiscale Biological Heterogeneity

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Riccardo Aucello,
  • Nicola Licheri,
  • Elena Rosso,
  • Giulio Ferrero,
  • Sandro Gepiro Contaldo,
  • Simone Pernice,
  • Pietro Lió,
  • Francesca Cordero,
  • Marco Beccuti

摘要

Biological phenomena are based on the precise and accurate cooperation of a non-random combination of molecules implicated in several pathways and networks. In the view of precision medicine, the plethora of omics data accrued sheds light on the comprehension of molecules cooperation. However, these data bring noise and redundancy that it is necessary to consider during the data analysis. A combination of omics data resources, integrated to parameterize mechanistic models, and multiphase Ensemble Feature Selection (EFS) is proposed. Through EFS, we characterized the metabolic heterogeneity of three distinct glycolysis-associated clusters (GACs) in colorectal cancer. Our study reveals that the EFS-derived genetic signatures associated with each GAC group also characterize three glycolysis profiles previously identified. GAC1 demonstrated unique separation, while GAC2 and GAC3 exhibited overlapping characteristics.