GATNNCDA: A Method Based on Graph Attention Network and Multi-Layer Neural Network for Predicting circRNA-Disease...

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Circular RNAs (circRNAs) are non-coding RNAs implicated in human diseases.
• Experimental identification of circRNA-disease interactions is costly and time-consuming.
• There is a critical need for efficient computational methods to predict these associations.

Purpose of the Study:

• To develop and validate a novel computational method, GATNNCDA, for predicting circRNA-disease associations.
• To leverage Graph Attention Network (GAT) and neural network (NN) models for enhanced prediction accuracy.
• To provide a reliable tool for discovering potential circRNA-disease relationships.

Main Methods:

• GATNNCDA integrates disease semantic similarity, circRNA functional similarity, and Gaussian Interaction Profile (GIP) kernel similarities.
• Initial node features are derived from integrated similarities within a heterogeneous circRNA-disease graph.
• Graph Attention Network (GAT) is employed for advanced feature extraction, followed by an NN-based classifier for prediction.

Main Results:

• GATNNCDA achieved high performance with an average AUC of 0.9613 and AUPR of 0.9433 on the CircR2Disease dataset.
• The method demonstrated superior performance compared to existing state-of-the-art approaches.
• Case studies confirmed a high proportion of predicted circRNA-disease associations for breast cancer and hepatocellular carcinoma.

Conclusions:

• GATNNCDA is an effective and reliable computational tool for discovering circRNA-disease associations.
• The proposed method offers a significant advancement in the field of circRNA-disease interaction prediction.
• GATNNCDA can accelerate research into the roles of circRNAs in human diseases.