ITRPCA: a new model for computational drug repositioning based on improved tensor robust principal component...

Area of Science:

• Bioinformatics
• Computational Biology
• Drug Discovery

Background:

• Drug repositioning accelerates the discovery of new uses for existing drugs.
• Computational methods offer cost-effective and efficient alternatives to experimental drug screening.
• Challenges in computational drug repositioning include sparse data, multi-source information, and noise.

Purpose of the Study:

• To develop an improved tensor robust principal component analysis (ITRPCA) for predicting drug-disease associations.
• To enhance the accuracy and efficiency of computational drug repositioning.
• To address data sparsity and noise in multi-source datasets.

Main Methods:

• Utilized a weighted k-nearest neighbor (WKNN) approach to increase data density.
• Constructed drug and disease tensors integrating multi-similarity matrices and updated association matrices.
• Applied ITRPCA with range constraints to identify low-rank tensors and noise for prediction.

Main Results:

• ITRPCA demonstrated higher prediction accuracy compared to five existing methods.
• The method exhibited remarkable computational efficiency.
• Cross-validation and independent testing confirmed the effectiveness of ITRPCA.

Conclusions:

• ITRPCA is a robust and efficient method for predicting drug-disease associations.
• The approach effectively handles multi-source data and noise in drug repositioning.
• Case studies validated the practical applicability of ITRPCA in identifying potential drug indications.