KGANCDA: predicting circRNA-disease associations based on knowledge graph attention network.

Area of Science:

• Biochemistry
• Genomics
• Computational Biology

Background:

• Circular RNAs (circRNAs) are increasingly recognized for their roles in disease pathogenesis.
• circRNAs show potential as biomarkers for disease diagnosis and treatment.
• Existing computational methods for predicting circRNA-disease associations suffer from data sparsity.

Purpose of the Study:

• To develop a novel computational method for predicting circRNA-disease associations.
• To address the limitations of existing methods, particularly data sparsity.

Main Methods:

• Constructed circRNA-disease knowledge graphs incorporating data on circRNA, disease, miRNA, and lncRNA.
• Employed a knowledge graph attention network (KGANCDA) to generate entity embeddings by weighing neighbor importance.
• Utilized a multilayer perceptron to predict circRNA-disease association scores based on learned embeddings.

Main Results:

• KGANCDA effectively captures both low-order and high-order neighbor information, mitigating data sparsity.
• Achieved superior performance compared to state-of-the-art methods in 5-fold cross-validation.
• Case studies confirmed KGANCDA's efficacy in predicting potential circRNA-disease associations.

Conclusions:

• KGANCDA represents an effective computational tool for predicting circRNA-disease associations.
• The method's ability to leverage multisource associations and high-order information enhances prediction accuracy.
• This approach holds promise for advancing the clinical application of circRNAs as biomarkers.