CFNCM: Collaborative filtering neighborhood-based model for predicting miRNA-disease associations.

Area of Science:

• Genomics
• Computational Biology
• Biomedical Informatics

Background:

• MicroRNAs (miRNAs) play a crucial role in human diseases.
• Understanding miRNA-disease interactions is vital for disease mechanism comprehension.
• miRNA-disease associations can serve as biomarkers and drug targets.

Purpose of the Study:

• To develop a computational model for predicting potential miRNA-disease associations.
• To overcome the limitations of expensive and time-consuming experimental methods.
• To enhance the identification of novel miRNA-disease links for diagnostic and therapeutic purposes.

Main Methods:

• Proposed the Collaborative Filtering Neighborhood-based Classification Model (CFNCM).
• Integrated miRNA and disease similarity matrices as input features.
• Employed user-based collaborative filtering to generate association scores and class labels (1 for positive association, 0 otherwise).
• Utilized machine learning algorithms, including Support Vector Machine (SVM), with 10-fold cross-validation and GridSearchCV for optimization.

Main Results:

• The CFNCM model demonstrated strong predictive performance.
• The SVM classifier achieved an Area Under the Curve (AUC) of 0.96.
• Validation on top 50 breast and lung cancer-related miRNAs confirmed 46 and 47 predicted associations in dbDEMC and miR2Disease databases, respectively.

Conclusions:

• The proposed CFNCM model is effective for predicting miRNA-disease associations.
• The model offers a cost-effective and efficient alternative to experimental methods.
• Accurate prediction of miRNA-disease links can significantly advance early detection, diagnosis, and treatment strategies for human disorders.