Integrative Multi-Omics and Computational Modeling for Biomarker Discovery in Complex Human Diseases

Complex human diseases—such as cancer, neurodegenerative disorders, autoimmune conditions, cardiometabolic disease, and chronic inflammatory syndromes—arise from interacting genetic, molecular, environmental, and lifestyle factors. This biological “cross-talk” often spans multiple molecular layers (genome, epigenome, transcriptome, proteome, metabolome, microbiome) and varies across tissues, time, and patient subgroups. As a result, single-omic biomarker approaches frequently fail to capture disease heterogeneity, predict progression, or guide treatment selection in a robust and reproducible way.



This Research Topic focuses on integrative multi-omics and computational modeling strategies that enable reliable biomarker discovery and validation in complex diseases. We welcome contributions that develop or apply methods to fuse multi-modal biological data with clinical phenotypes, imaging, digital health, and real-world evidence to identify biomarkers that are predictive, mechanistically grounded, and clinically actionable. Particular emphasis is placed on approaches that move beyond correlation to capture causal structure, pathways, and disease dynamics—supporting patient stratification, early detection, prognosis, therapy response prediction, and monitoring.



We encourage submissions across a broad spectrum of computational paradigms, including (but not limited to) machine learning and deep learning, network biology, mechanistic and systems biology models, Bayesian and probabilistic frameworks, causal inference, and hybrid models that combine mechanistic insight with data-driven prediction. Studies addressing reproducibility, interpretability, batch effects, missingness, cohort bias, fairness, and cross-cohort generalization are especially welcome, as are benchmark datasets and open tools that improve transparency and reuse.



Suggested themes and subtopics include:



· Multi-omics integration methods (horizontal/vertical integration; single-cell and spatial multi-omics; longitudinal data)



· Biomarker discovery for diagnosis, prognosis, and treatment response (including multi-marker panels and composite scores)



· Disease subtyping and patient stratification using multi-modal data (omics + clinical + imaging + digital biomarkers)



· Causal and mechanistic modeling to link biomarkers to pathways, disease drivers, and therapeutic targets



· Network-based biomarker discovery (gene regulatory networks, protein interaction networks, pathway activity inference)



· AI/ML approaches that emphasize interpretability and clinical translation (explainable AI, uncertainty estimation)



· Validation frameworks (external validation, prospective studies, assay transferability, clinical utility evaluation)



· FAIR, reproducible pipelines; federated and privacy-preserving analytics; robust harmonization across cohorts



By bringing together methodological innovation and disease-focused applications, this Research Topic aims to advance biomarker discovery from “signals in data” to validated markers with biological meaning and clinical value, accelerating precision medicine for complex human diseases.