LITERATURE REVIEW OF PREDICTING DISEASE GENE ASSOCIATION

Computer Engineering

Disease-gene interaction prediction plays an important role in understanding diseases and supporting drug discovery. This project uses Graph Autoencoders (GAEs) to model these interactions by analysing biological networks. GAEs are effective for this task because they can capture complex graph structures and learn meaningful representations of relationships between genes and diseases. In this approach, a graph is constructed where nodes represent genes and diseases, and edges represent known interactions between them. A Graph Convolutional Network (GCN)-based encoder is used to learn structural features from the graph, while a decoder reconstructs the graph to identify potential new interactions. The model is trained by minimizing reconstruction loss, making it suitable for handling large-scale datasets. Future work will focus on improving model interpretability and incorporating domain-specific knowledge to enhance prediction performance.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.