Scholarlink

Background <p>Untargeted metabolomics has emerged as a powerful approach to uncover metabolic dysregulation associated with cancer progression. When integrated with a machine learning strategy it facilitates the discovery of key metabolic pathways and predictive biomarkers with high diagnostic and prognostic value.</p> Methods <p>In this study, we employed liquid chromatography coupled to high-resolution Tribrid Orbitrap mass spectrometry to perform comprehensive metabolic profiling of bladder cancer (BLCA) as well as predict invasiveness of the disease.</p> Results <p>By leveraging both in-house retention time-based MS/MS spectral libraries and commercial databases, we robustly identify over 2000 metabolites. In addition, this platform allows identification of novel pathways highlighting metabolic vulnerabilities in BLCA. The application of machine learning algorithms and advanced computational modeling uncovered metabolic signatures that differentiate BLCA from adjacent normal/benign samples and distinguish muscle-invasive from non-muscle-invasive bladder cancer. Our integrative analytical pipeline addresses key challenges in metabolomics-including high dimensionality, metabolite annotation, and biological variability-through feature selection and predictive modeling. We identify candidate metabolic markers with strong potential for early detection and characterize invasiveness of the disease and identify potential therapeutic target pathways.</p> Conclusions <p>This work highlights the power of combining untargeted metabolomics with machine learning to map the metabolic landscape of BLCA and to accelerate the development of precision diagnostics and future 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

Leveraging untargeted metabolomics in combination with machine learning to uncover novel insights into bladder cancer

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月1日

查看全文
  • Abu Hena Mostafa Kamal,
  • Vasanta Putluri,
  • Tanmay Gandhi,
  • Chandra Shekar R. Ambati,
  • Chandra Sekhar Amara,
  • Karthik Reddy Kami Reddy,
  • Meredith Lauren Spradlin,
  • Amrit Koirala,
  • Sachin B. Jorvekar,
  • Sandra L. Grimm,
  • Dexue Fu,
  • Krishna Parsawar,
  • Felice de Jong,
  • Chris Beecher,
  • Subrata Sen,
  • Seth P. Lerner,
  • M. Minhaj Siddiqui,
  • Yair Lotan,
  • Livia S. Eberlin,
  • Arun Sreekumar,
  • Cristian Coarfa,
  • Nagireddy Putluri

摘要

Background

Untargeted metabolomics has emerged as a powerful approach to uncover metabolic dysregulation associated with cancer progression. When integrated with a machine learning strategy it facilitates the discovery of key metabolic pathways and predictive biomarkers with high diagnostic and prognostic value.

Methods

In this study, we employed liquid chromatography coupled to high-resolution Tribrid Orbitrap mass spectrometry to perform comprehensive metabolic profiling of bladder cancer (BLCA) as well as predict invasiveness of the disease.

Results

By leveraging both in-house retention time-based MS/MS spectral libraries and commercial databases, we robustly identify over 2000 metabolites. In addition, this platform allows identification of novel pathways highlighting metabolic vulnerabilities in BLCA. The application of machine learning algorithms and advanced computational modeling uncovered metabolic signatures that differentiate BLCA from adjacent normal/benign samples and distinguish muscle-invasive from non-muscle-invasive bladder cancer. Our integrative analytical pipeline addresses key challenges in metabolomics-including high dimensionality, metabolite annotation, and biological variability-through feature selection and predictive modeling. We identify candidate metabolic markers with strong potential for early detection and characterize invasiveness of the disease and identify potential therapeutic target pathways.

Conclusions

This work highlights the power of combining untargeted metabolomics with machine learning to map the metabolic landscape of BLCA and to accelerate the development of precision diagnostics and future therapeutic strategies.