Extra Trees Method for Predicting LncRNA-Disease Association Based On Multi-Layer Graph Embedding Aggregation.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Long non-coding RNAs (lncRNAs) are increasingly recognized for their roles in various diseases.
• Accurate identification of lncRNA-disease associations is crucial for developing novel therapeutic strategies.
• Existing computational methods for predicting these associations have limitations.

Purpose of the Study:

• To propose an accurate computational method, MLGCNET, for discovering potential long non-coding RNA-disease associations.
• To improve the prediction performance compared to existing state-of-the-art methods.

Main Methods:

• Reconstruction of lncRNA and disease similarity networks using top k similar information.
• Construction of a lncRNA-disease heterogeneous network (LDN).
• Application of Multi-Layer Graph Convolutional Network (MLGCN) on LDN for latent feature extraction.
• Utilizing Extra Trees algorithm to calculate the probability of lncRNA-disease association.

Main Results:

• MLGCNET demonstrated superior prediction performance in extensive 5-fold cross-validation experiments.
• The model outperformed existing state-of-the-art methods in identifying lncRNA-disease associations.
• Case studies validated the model's effectiveness for specific diseases.

Conclusions:

• MLGCNET is an effective and practical tool for predicting potential lncRNA-disease associations.
• The proposed method offers a valuable approach for advancing disease therapy through lncRNA research.
• The study highlights the potential of graph convolutional networks in biological network analysis.