X-omics ACTION demonstrator multi-omics analysis workflow

X-omics ACTION demonstrator multi-omics analysis workflow performs multi-omics integration and analysis to identify shared patterns across genome-wide DNA methylation, targeted metabolomics, and behavioral data and to relate omics-derived latent features to phenotypic dimensions such as childhood externalizing behavior.

Key Features:

• Modular Workflow: Implemented in Nextflow to provide a modular and scalable computational structure for processing diverse omics datasets.
• FAIR Practices: Incorporates FAIR-aligned elements including version control, appropriate licensing, and rich semantic metadata to support findability, interoperability, and reusability.
• Multi-Omics Integration: Processes genome-wide DNA methylation, targeted metabolomics, and behavioral data for combined analyses across omics layers.
• Analytical Methods: Employs integrative and unsupervised methods such as Similarity Network Fusion (SNF) for clustering and Multi-Omics Factor Analysis (MOFA) and Multiple Correspondence Analysis (MCA) for latent-dimension discovery.
• Correlation and Visualization: Enables integration-driven correlation analyses between omics-derived features and phenotypic dimensions, with downstream results visualization.

Scientific Applications:

• Multi-layer pattern discovery: Identify shared molecular signatures across DNA methylation and metabolomics layers linked to behavioral measures.
• Interplay analysis: Study the interplay between genetic/epigenetic, metabolic, and behavioral factors in cohort-based research.
• Phenotype association: Relate omics-derived latent features to phenotypic dimensions, including investigation of mechanisms underlying childhood externalizing behavior.

Methodology:

Implemented in Nextflow; performs pre-processing of multi-omics data, clustering with Similarity Network Fusion (SNF), dimensionality reduction with MOFA and MCA, integration and correlation analyses between omics and phenotypic traits, and result visualization.