AMPGLDA: Predicting LncRNA-Disease Associations Based on Adaptive Meta-Path Generation and Multi-Layer Perceptron

Increased evidence suggests that long non-coding RNAs (lncRNAs) hold a vital position in intricate human diseases. Nonetheless, the current pool of identified lncRNAs linked to diseases remains restricted. Hence, a reliable and cost-effective computational method is needed to predict probable correlations between lncRNAs and diseases, which could help reveal underlying mechanisms and foster the development of novel treatments. In this study, we propose a novel approach for predicting the associations between lncRNAs and diseases, which relies on the adaptive meta-path generation and multi-layer perceptron (AMPGLDA). Firstly, we integrate information about lncRNA, diseases, and miRNAs to construct a heterogeneous graph. Then, we utilize principal component analysis to extract global features from nodes. Based on this heterogeneous graph, AMPGLDA adaptively generates multiple meta-path graph structures and uses a graph convolutional neural network to learn the semantic feature representations of lncRNA and disease from the meta-path. Ultimately, AMPGLDA utilizes a deep neural network classifier to accurately predict the association between lncRNA and disease. Cross-validation experiments revealed that AMPGLDA outperformed 5 other cutting-edge prediction methods. Moreover, ablation studies and case studies confirmed its technical contributions and its capability to identify disease related lncRNAs.