A Useful Tool for the Identification of DNA-binding Proteins Using Graph Convolutional Network

Both DNAs and proteins are important components of living organisms. DNA-binding proteins are a kind of helicase, which is a protein specifically responsible for binding to DNA single stranded regions. It plays a key role in the function of various biomolecules. Although there are some prediction methods for the DNA-binding proteins sequences, the use of graph neural networks in this research is still limited. In this article, using graph neural networks, we developed a novel predictor GCN-DBP for protein classification prediction. Each protein sequence is treated as a document in this study, and then document is segmented according to the concept of k-mer. This research aims to use document word relationships and word co-occurrence as a corpus to construct a text graph. Then, the predictor learns protein sequence information by two-layer graph convolutional networks. In order to compare the proposed method with other four existing methods, we have conducted more experiments. Finally, we tested GCN-DBP on the independent data set PDB2272. Its accuracy reached 64.17% and MCC reached 28.32%. The results show that the proposed method is superior to the other four methods and will be a useful tool for protein classification.