Automated computational platform for interpretation of multi-omics data reveals novel disease biomarkers

Introduction: This study presents an automated computational platform for multi-omics integration and analysis through several case studies, incorporating transcriptomics, epigenetics, and metabolomics. By leveraging robust statistical modeling, bioinformatics and machine learning, the platform enables comprehensive understanding and novel insights into disease physiology. Hypothesis: We hypothesize that integrating multi-omics data using computational modeling can uncover previously unknown molecular interactions, regulatory networks, and biomarkers, ultimately advancing our understanding of disease mechanisms and informing therapeutic strategies. Methods: The platform processes data using streamlined and standardized pipelines, ensuring reproducibility. Quality assessment, normalization, and statistical methods are applied for data analysis. Associations inferred using mixed-effects linear modeling, with adjustments for confounding variables and batch effects, uncovered across omic layers are used to train and evaluate highly accurate machine learning models for functional discovery. Data: The platform has been applied to study diverse datasets, including transcriptomics, proteomics, metabolomics, epigenomics and genomics. Results: Our results demonstrate that our automated platform serves as a unique computational biology tool, capable of automatically transforming multi-omics data into actionable functional discoveries. It integrates diverse omics datasets, employs advanced statistical methods and machine learning to simplify complex analyses. This automated approach saves time, reduces complexity, and empowers researchers to make better-informed decisions. The platform successfully identified novel molecular biomarkers, regulatory pathways, and dynamic network interactions contributing to disease progression. Importantly, its simplicity ensures that researchers without extensive computational expertise can uncover groundbreaking insights, enabling functional discoveries at the systems biology level. Conclusions: This work underscores the potential of automated computational modeling and bioinformatics to transform research. The platform provides a robust solution for multi-omics data analysis and interpretation, paving the way for improved disease modeling, personalized medicine strategies and ultimately improved patient outcomes. Private funds This abstract was presented at the American Physiology Summit 2025 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.