Drug Combination Recommendation Model for Systemic Lupus Erythematosus and Antiphospholipid Syndrome

Area of Science:

• Computational biology
• Pharmacogenomics
• Clinical informatics

Background:

• Systemic Lupus Erythematosus (SLE) and Antiphospholipid Syndrome (APS) are autoimmune disorders requiring careful drug selection.
• Drug-drug interactions and patient variability pose significant risks in managing these conditions.

Purpose of the Study:

• To develop a personalized drug recommendation model for SLE and APS patients.
• To mitigate risks associated with drug regimens by considering combination frequency, risk assessment, and genetic interactions.

Main Methods:

• Extracted drug combination and gene interaction data from MIMIC-III, Drug Bank, and Gene Expression Omnibus.
• Developed a greedy algorithm to evaluate drug combinations, risk levels, and gene interactions.
• Proposed a model for personalized, low-risk drug recommendations.

Main Results:

• The model effectively reduces potential drug-related risks while maintaining therapeutic effects.
• Demonstrated robust performance across various drug combinations and clinical scenarios.
• Provided effective drug substitution suggestions for clinicians.

Conclusions:

• The study offers a theoretical and technical foundation for personalized therapy and precision medicine in autoimmune disorders.
• Highlights the potential of computational models in optimizing drug regimens for complex conditions like SLE and APS.