Related Experiment Video
Updated: Jul 16, 2026
07:54
Assessing Hepatic Metabolic Changes During Progressive Colonization of Germ-free Mouse by 1H NMR Spectroscopy
Published on: December 15, 2011
13.1K
Integrating HRMAS-NMR Data and Machine Learning-Assisted Profiling of Metabolite Fluxes to Classify Low- and
Safia Firdous1,2, Zubair Nawaz3, Rizwan Abid1
1Biopharmaceuticals and Biomarkers Discovery Lab, School of Biochemistry and Biotechnology, University of the Punjab, Lahore, 54590, Pakistan.
Interdisciplinary Sciences, Computational Life Sciences
|September 27, 2024
Summary
This study used metabolomics and machine learning to identify metabolic biomarkers for classifying glioma tumor grades. These findings show promise for improving diagnosis and treatment planning for brain tumors.
Area of Science:
- Neuro-oncology
- Metabolomics
- Machine Learning
Background:
- Gliomas, including glioblastomas, are aggressive central nervous system tumors that are challenging to diagnose and classify.
- Metabolomics and magnetic resonance spectroscopy (MRS) offer potential for in vivo and ex vivo tumor grading.
Purpose of the Study:
- To investigate tissue-based metabolic signatures for classifying low- and high-grade gliomas.
- To identify potential metabolic biomarkers for glioma classification.
Main Methods:
- Analysis of 46 glioma tumor samples using high-resolution magic angle spinning nuclear magnetic resonance (HRMAS-NMR) spectroscopy.
- Integration of machine learning (ML) algorithms to identify discriminative spectral regions.
- Validation using statistical analyses and HRMAS-NMR of paired plasma samples.
Main Results:
- Machine learning models achieved high accuracy in sub-classifying gliomas (up to 91%).
- 40 spectral regions corresponding to 18 metabolites were identified as potential biomarkers.
- N-acetyl aspartate, glutamate, and glutamine were identified as key markers, validated in plasma samples.
Conclusions:
- HRMAS-NMR combined with ML can effectively classify glioma grades using metabolic signatures.
- Identified metabolic biomarkers hold promise for targeted treatment planning and clinical interventions in neuro-oncology.
