OmniPath: integrated knowledgebase for multi-omics analysis

Area of Science:

• Bioinformatics
• Systems Biology
• Computational Biology

Background:

• Omics data analysis relies heavily on prior biological knowledge.
• Existing knowledge resources are often fragmented and difficult to access for computational methods.

Purpose of the Study:

• To develop OmniPath, a comprehensive database integrating diverse molecular interactions and biomolecule information.
• To create accessible tools, including OmniPath Explorer and client packages, for utilizing this knowledge in omics analysis.

Main Methods:

• Integrated data from 168 diverse resources, including literature-curated interactions, predictions, and large-scale databases.
• Developed OmniPath Explorer with a large language model agent for interactive knowledge browsing.
• Created Python and R/Bioconductor client packages and a Cytoscape plugin for seamless integration with omics analysis environments.

Main Results:

• OmniPath consolidates causal protein-protein, gene regulatory, microRNA, and enzyme-post-translational modification interactions.
• The database includes information on cell-cell communication, protein complexes, and biomolecule functions, localization, and structure.
• OmniPath Explorer and associated packages facilitate customized prior knowledge retrieval for omics studies.

Conclusions:

• OmniPath provides a unified and accessible resource for prior biological knowledge, crucial for omics data interpretation.
• The developed tools enable the application of integrated knowledge for mechanistic and causal modeling in bulk, single-cell, and spatial multi-omics data analysis.