AI and ML in Untargeted Metabolomics and Exposomics:
Metabolomics uses high-throughput approaches to measure a range of metabolites and small molecules in biological samples, providing critical insights into human health and disease. One application, untargeted metabolomics, allows for unbiased global analysis of the metabolome to identify key metabolites that contribute to or are indicative of health conditions. Recent advances in AI and ML have significantly enhanced untargeted metabolomics workflows, especially in the context of high-resolution mass spectrometry (HRMS) exosomes. This emerging field detects endogenous metabolites and exogenous chemicals in human tissues and correlates environmental exposures with disease outcomes. AI and ML applications have improved data quality, rigor, detection, and identification of chemicals, facilitating screening and diagnostic outcomes for major diseases.
Metabolism is the process by which the body produces essential metabolites, including catabolism (breaking down molecules to obtain energy) and anabolism (synthesis of compounds needed by cells). Metabolomics captures endogenous metabolites and signaling molecules involved in gene expression, protein function, and enzyme activity. Targeted metabolomics measures specific metabolites, while untargeted metabolomics provides broader semi-quantitative analysis of thousands of small molecules. This holistic approach, called exposomics, incorporates environmental exposures, diet, lifestyle, and psychosocial factors to illuminate their impact on health. While much remains unknown about the human exposome, AI and ML are making advances in the discovery and analysis of these complex datasets to advance our understanding of chemical exposures and their impact on human health.
Untargeted Metabolomics Workflow:
When analyzing biological matrices such as serum, plasma, and urine, the untargeted metabolomics workflow typically involves the separation of complex mixtures using LC or GC column chromatography followed by detection and measurement by HRMS. The process includes sample preparation, data acquisition, pre- and post-processing, data analysis, and chemical identification. Metabolites and chemicals are extracted using organic solvents and analyzed by HILIC or reversed-phase chromatography in the case of LC, or derivatized for GC analysis. HRMS generates data in three dimensions: mass-to-charge ratio, retention time, and abundance. AI and ML tools play a key role in data processing, feature selection, and chemical identification to enhance the analysis of metabolomics data and its biological interpretation.
Data processing in untargeted metabolomics:
Metabolomics raw data is complex due to linear and nonlinear interactions between metabolites and challenges in mass spectrometry data structure. Preprocessing is crucial in converting 3D data from LC-MS into 2D aligned peak tables required for downstream analysis. Algorithms such as XCMS, MZmine, and MS-Dial are used for preprocessing, but only a few methods are widely accepted. Recent developments include quality control measures and new peak selection algorithms such as CPC and Finnee that enhance peak selection. Machine learning tools such as WiPP, MetaClean, Peakonly, NeatMS, NPFimg, and EVA promise to improve the accuracy and reliability of data processing.
AI and ML in Biomarker Discovery:
Traditional univariate and multivariate models perform multiple hypothesis testing to identify metabolite signatures associated with phenotypes, but require assistance with the correlation structure of metabolomic data. AI and ML methods address these limitations by building and testing models directly on the data to uncover relationships between phenotypes, exposures, and diseases. Tools such as LASSO, PCA, HCA, SOM, PLS-DA, RF, and newer techniques such as ANN and DL have enabled the identification of important biomarkers and metabolite signatures. AI and ML have been used to detect diseases such as NAFLD, COVID-19, Alzheimer's, and depression, demonstrating their potential in metabolomic research.
Metabolite Identification in Biomarker Discovery:
Metabolite identification is essential for biomarker discovery and requires annotating selected peaks using metabolite databases and spectral libraries such as GNPS, Metlin, and Human Metabolome Database. This process requires matching m/z and MS/MS fragmentation data to confirm metabolites. Despite available databases, the spectral match rate for specialized chemicals still needs to be high. Advances in cognitive metabolomics using ML and NLP and in silico tools such as CSI: FingerID and CFM-ID have improved the identification accuracy. Expanding spectral libraries and developing new annotation tools are essential to better identify and understand both endogenous and exogenous chemicals.
Advances in non-targeted chemical analysis:
Advances in untargeted chemical analysis and AI/ML tools have significantly reduced costs and enabled large-scale studies. AI/ML aids in data extraction, mining, and annotation, which are essential for biomarker discovery. A major challenge is the annotation of unknown metabolites, which are essential for biological interpretation. There has been a focus on developing experimental databases and AI/ML models to enhance metabolite identification. However, current algorithms often miss low-concentration chemicals, indicating the need for improved ML classifiers. Integrating biology-driven approaches with measurement-based methods may reveal unknown chemicals that impact health and drive discoveries on exosomes and precision health.
Please check paper. All credit for this research goes to the researchers of this project. Also, don't forget to follow us. twitter And our Telegram Channel and LinkedIn GroupsUp. If you like our work, you will love our Newsletter..
Please join us 47,000+ ML subreddits
Check out our upcoming AI webinars here
Sana Hassan, a Consulting Intern at Marktechpost and a dual degree student at Indian Institute of Technology Madras, is passionate about applying technology and AI to address real-world challenges. With a keen interest in solving practical problems, she brings a fresh perspective to the intersection of AI and real-world solutions.
