MuTATE-an R package for comprehensive multi-objective molecular modeling.

Area of Science:

• Computational biology
• Bioinformatics
• Machine learning

Background:

• Multi-omics studies advance precision medicine but challenge current machine learning models due to limited interpretability.
• Existing tools struggle to simultaneously identify disease subgroups, biomarkers, and explain multiple clinical endpoints.
• Current approaches require advanced computational skills and extensive manual interpretation.

Purpose of the Study:

• To develop an automated, comprehensive machine learning approach for molecular modeling.
• To enable simultaneous identification of disease subgroups and their defining molecular biomarkers.
• To provide an interpretable model explaining multiple clinical endpoints for precision medicine.

Main Methods:

• Developed Multi-Target Automated Tree Engine (MuTATE) for automated decision tree construction.
• Incorporated multiple targets and target weights for multi-objective modeling.
• Enabled user-friendly visualization of relationships between molecular biomarkers and patient subgroups.

Main Results:

• MuTATE facilitates automated and comprehensive molecular modeling with interpretable decision trees.
• The tool visualizes relationships between molecular biomarkers and patient subgroups across multiple clinical endpoints.
• MuTATE enhances efficiency by eliminating the need for manual synthesis of multiple non-explainable models.

Conclusions:

• MuTATE offers a flexible and versatile solution for complex disease modeling, including cancer.
• The tool has the potential to transform biomarker discovery and subtype identification for personalized treatment strategies.
• MuTATE advances understanding of disease mechanisms and improves therapeutic decisions in precision medicine.