Microbe-disease associations prediction by graph regularized non-negative matrix factorization with L 2 , 1 $$...

Area of Science:

• Microbiology
• Computational Biology
• Bioinformatics

Background:

• Microbes play crucial roles in biological processes and disease pathogenesis.
• Identifying microbe-disease associations is vital for developing biomarkers and therapeutic targets for complex human diseases.
• Traditional experimental methods for microbe-disease association are costly and time-consuming.

Purpose of the Study:

• To introduce iPALM-GLMF, a novel computational method for predicting microbe-disease associations.
• To improve the accuracy and interpretability of microbe-disease association prediction.
• To provide an efficient alternative to experimental methods.

Main Methods:

• Modeled microbe-disease association prediction as non-negative matrix factorization with graph dual regularization and L1 norm regularization.
• Employed non-negative double singular value decomposition for initialization.
• Utilized an inertial Proximal Alternating Linear Minimization iterative process for optimization.

Main Results:

• iPALM-GLMF demonstrated superior performance compared to existing methods in leave-one-out and fivefold cross-validation.
• Case studies confirmed the method's effectiveness in predicting potential microbial-disease associations.
• The method enhances sparsity and interpretability of feature matrices.

Conclusions:

• iPALM-GLMF is an effective and accurate computational tool for predicting microbe-disease associations.
• The method offers a valuable approach for identifying potential microbial biomarkers and drug targets.
• The developed model is publicly available for broader research application.