Approaches to Integrating Metabolomics and Multi-Omics Data: A Primer.

Area of Science:

• Biochemistry and Bioinformatics
• Systems Biology

Background:

• Metabolomics studies biochemical reactions and their response to perturbations, reflecting genomic, transcriptomic, and proteomic information.
• It is the omics layer closest to the phenome, offering insights into biological functions and mechanisms.
• Integrating metabolomics with other omics data (genomics, transcriptomics, proteomics) enhances biological understanding and reveals complex associations.

Purpose of the Study:

• To review and propose a classification system for statistical multi-omics data integration approaches.
• To provide guidance on selecting appropriate integrative analysis pipelines.
• To specifically assist researchers working with metabolomics and genomics data.

Main Methods:

• Review of existing computational tools for multi-omics data integration.
• Proposal of five classification criteria: hypothesis, data types, strategies, study design, and study focus.
• Categorization of statistical multi-omics integration methods into distinct classes.

Main Results:

• A framework is presented to classify statistical multi-omics data integration methods.
• The proposed criteria enable a systematic approach to understanding and selecting integration strategies.
• The review highlights the importance of considering specific aspects when choosing an analysis pipeline.

Conclusions:

• The classification system aids researchers in navigating the complex landscape of multi-omics data integration.
• Choosing the right statistical pipeline is crucial for effectively combining metabolomics with other omics data.
• This work supports researchers, particularly those new to metabolomics and genomics integration, in their analytical choices.