Mechanism-Aware Machine Learning in Metabolomics and Clinical Translation: Signatures, Pathways, and Patient Impact

Dear Colleagues,

This Special Issue invites contributions at the intersection of clinical metabolomics and machine learning (ML), highlighting how data-driven methods are advancing the discovery, validation, and translation of metabolic biomarkers into tools for diagnosis, prognosis, and therapeutic decision-making. We seek studies that cover the entire pipeline, from raw data to clinically actionable models, emphasizing methodological rigor, biological interpretability, and reproducibility in clinical settings.

We particularly welcome studies that leverage LC/GC-MS, MS/MS, NMR, and imaging mass spectrometry to generate high-quality, harmonized datasets across sites and instruments.

Key topics include feature detection, metabolite identification, spectral library expansion, and graph-based or Bayesian approaches for chemical inference. We are particularly interested in studies that integrate FAIR data practices (e.g., mzML, MetaboLights) and methods like interpretable ML models (e.g., sparsity-inducing models, SHAP analysis), deep learning, self-supervision, and federated learning for multi-center studies. Contributions exploring causal inference and counterfactual frameworks for treatment-effect estimation are also encouraged.

Clinical applications may involve biomarker discovery, risk prediction, minimal residual disease monitoring, and pharmaco-metabolomics for response and toxicity forecasting. We are also interested in the longitudinal modeling of metabolic trajectories with the integration of multi-omics data (genomics, transcriptomics, proteomics, microbiome, and wearable data), and mechanistic studies linking learned signatures to metabolic pathways. Studies that employ Machine Learning to elucidate disease biology (e.g., inflammation, mitochondrial dysfunction) or to quantify therapy-induced metabolic changes are within the scope of this Special Issue, especially those demonstrating clinical utility through external validation, decision-curve analysis, and health-economic impact.

Submissions should provide transparent reporting of data preprocessing, model development (training/validation/test splits, cross-site generalization), bias and confounder controls (age, sex, ancestry, medications, diet), and sensitivity analyses.

We encourage sharing open, reusable code and pipelines (e.g., Snakemake, Nextflow) and documenting model cards to promote replication and regulatory compliance.

By integrating advances in computational methodology and clinical translation, this Special Issue aims to define best practices for ML-enabled metabolomics, highlight biologically grounded and interpretable models, and accelerate the development of clinically deployable metabolic biomarkers and decision support systems. We seek studies that deliver validated, generalizable tools capable of improving patient care from bench to bedside.

Dr. Vasuk Gautam
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Metabolites is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2700 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• bioinformatics
• machine learning
• metabolomics
• integrative omics
• multi-omics
• translational research
• data analysis
• clinical research