Shi H, Feng H, Lu Z, et al. GRA-GCN: Dense Granule Protein Prediction in Apicomplexa Protozoa Through Graph Convolutional Network. IEEE/ACM Trans Comput Biol Bioinform. 2022;20(3):1963-1970. https://doi.org/10.1109/TCBB.2022.3224836.
This study presents GRA-GCN, a novel computational method for predicting dense granule proteins (GRAs) in Apicomplexa protozoa using a graph convolutional network (GCN). GRAs are linked to numerous parasitic diseases in farm animals, making accurate prediction essential for disease prevention and treatment. The model treats GRA prediction as a node classification problem, leveraging k-nearest neighbor algorithms to build feature graphs. Evaluated through 5-fold cross-validation, GRA-GCN outperformed several classic machine learning methods and state-of-the-art models. A web server has been developed to facilitate the use of this model in further research and applications.
Comments