iPiDA-GCN: Identification of piRNA-disease associations based on Graph Convolutional Network

Abstract

Motivation Piwi-interacting RNAs (piRNAs) play a critical role in the progression of various diseases. Accurately identifying the associations between piRNAs and diseases is important for diagnosing and prognosticating diseases. Although some computational methods have been proposed to detect piRNA-disease associations, it is challenging for these methods to effectively capture nonlinear and complex relationships between piRNAs and diseases because of the limited training data and insufficient association representation. Results With the growth of piRNA-disease association data, it is possible to design a more complex machine learning method to solve this problem. In this study, we propose a computational method called iPiDA-GCN for piRNA-disease association identification based on graph convolutional networks (GCNs). The iPiDA-GCN predictor constructs the graphs based on piRNA sequence information, disease semantic information and known piRNA-disease associations. Two GCNs (Asso-GCN and Sim-GCN) are used to extract the features of both piRNAs and diseases by capturing the association patterns from piRNA-disease interaction network and two similarity networks. GCNs can capture complex network structure information from these networks, and learn discriminative features. Finally, the full connection networks and inner production are utilized as the output module to predict piRNA-disease association scores. Experimental results demonstrate that iPiDA-GCN achieves better performance than the other state-of-the-art methods, benefitted from the discriminative features extracted by Asso-GCN and Sim-GCN. The iPiDA-GCN predictor is able to detect new piRNA-disease associations to reveal the potential pathogenesis at the RNA level. The data and source code are available at http://bliulab.net/iPiDA-GCN/.

IGCNSDA: unraveling disease-associated snoRNAs with an interpretable graph convolutional network

“…iPiDA_GCN [ 38 ] serves as a computational technique in this study, aimed at discerning piRNA–disease associations by harnessing the capabilities of GCN. It effectively extracts unique features from both piRNAs and diseases while leveraging association patterns within various networks.…”

piRNAs in the human retina and retinal pigment epithelium reveal a potential role in intracellular trafficking and oxidative stress

“…2A). These four piRNAs were found to have association with neuronal diseases, including Alzheimers' disease 42 and autism spectrum disorder. 43,44 Among the evaluated piRNAs, piR-hsa-1207 and piR-hsa-24683 were expressed more in the retina compared to RPE, which was in accordance with their CpM values from the RNA-seq data.…”

Databases and computational methods for the identification of piRNA-related molecules: A survey

“… The training time of GrowNet is long, which may affect the efficiency of the model. Alzheimer’s disease and Parkinson’s disease 14 iPiDA-GCN [34] GCN Disease semantic similarity and piRNA sequence similarity 10,149 19 11,981 0.7149 Designing two GCN modules (Asso-GCN and Sim-GCN) to extract information from piRNA-disease association network and two similarity networks respectively, which can enhance the diversity and robustness of the representation. Three different networks need to be built, which may cause data sparsity or incompleteness problems, and also increase the computational complexity of the model.…”

“…iPiDA-LTR [33] predicted that piR-hsa-23210 and piR-hsa-23209 were associated with multiple diseases, and their target genes played important roles in human spermatogenesis and nervous system development and may be key therapeutic targets. iPiDA-GCN [34] predicted that piR-hsa-31280 and piR-hsa-8245 were associated with cardiovascular diseases, and they were abnormally expressed in cardiovascular disease tissues and may be related to cardiac function and repair.…”