CAT Bridge: an efficient toolkit for gene-metabolite association mining from multiomics data

Area of Science:

• Multiomics data analysis
• Systems biology
• Bioinformatics

Background:

• Advancements in sequencing and mass spectrometry enable multiomics data acquisition.
• Challenges persist in associating gene-metabolite pairs due to complex biological networks.
• Nonlinear interactions, feedback loops, and time-dependent mechanisms complicate traditional analysis.

Purpose of the Study:

• Introduce Compounds And Transcripts Bridge (CAT Bridge), a platform for longitudinal multiomics analysis.
• Facilitate efficient identification of transcripts associated with metabolites using time-series omics data.
• Benchmark statistical methods for gene-metabolite association, including causality estimation and correlation.

Main Methods:

• Developed CAT Bridge, a user-friendly platform for longitudinal multiomics analysis.
• Integrated a range of statistical methods for evaluating gene-metabolite pair associations.
• Incorporated an artificial intelligence agent to assist in interpreting association results.

Main Results:

• CAT Bridge successfully identified genes involved in capsaicin biosynthesis in Capsicum chinense.
• Demonstrated the superior performance of the convergent cross-mapping method over traditional approaches in longitudinal multiomics.
• Validated the platform's utility with human, Escherichia coli, and C. chinense time-series datasets.

Conclusions:

• CAT Bridge simplifies access to established longitudinal multiomics analysis methods.
• Enables researchers to rapidly identify gene-metabolite pairs for further experimental validation.
• Facilitates a deeper understanding of complex biological systems through multiomics data integration.