Microbial Risk Score for Capturing Microbial Characteristics, Integrating Multi-omics Data, and Predicting Disease...

Area of Science:

• Microbiome research
• Computational biology
• Disease prediction

Background:

• Microbiome-wide association studies generate vast data for disease research.
• Unique microbiome data features hinder direct application in disease prediction.
• Developing robust methods is crucial for leveraging microbiome data in healthcare.

Approach:

• Proposed a microbial risk score (MRS) framework inspired by polygenic risk scores.
• MRS algorithm identifies disease-associated microbial taxa and integrates them into a continuous score.
• Developed a multi-omics data integration method for joint modeling with MRS.

Key Points:

• Evaluated MRS framework using COVID-19, GMHI, and type 1 diabetes cohorts.
• Created and validated disease-specific MRS for colorectal adenoma, colorectal cancer, Crohn's disease, and rheumatoid arthritis.
• Crohn's disease MRS achieved high AUCs (0.88 discovery, 0.86 validation).

Conclusions:

• The MRS framework demonstrates the utility of microbiome data for disease prediction.
• Highlights potential for multi-omics integration in understanding disease.
• Offers promising applications in disease diagnosis and prognosis.