Graph Neural Networks (GNNs) show great power in Knowledge Graph Completion (KGC) as they can handle non-Euclidean graph structures and do not depend on the specific shape or topology of the graph. However, many current GNN-based KGC models have difficulty in effectively capturing and utilizing the substantial structure and global semantic information in Knowledge Graphs (KGs). For more effective use of GNN for KGC, we innovatively propose the Semantic Similarity-based Interaction Graph Attention Network (SemSI-GAT) for the KGC task. In SemSI-GAT, we utilize BERT, a pre-trained language model, to learn the global semantic information and obtain semantic similarity between entities and their neighbors. Furthermore, we creatively design a novel encoder network called the interaction graph attention network and introduce a semantic similarity sampling mechanism to optimize the aggregation of interaction information between neighbors. By aggregating local features with interaction features, this network can generate more expressive structural embeddings. This network generates more expressive embeddings by fusing global semantic information, local structure features, and interaction features. The experimental evaluations demonstrate that the proposed SemSI-GAT outperforms existing state-of-the-art KGC methods on four benchmark datasets.

Protein-protein interactions (PPIs) are a major component of the cellular biochemical reaction network. Rich sequence information and machine learning techniques reduce the dependence of exploring PPIs on wet experiments, which are costly and time-consuming. This paper proposes a PPI prediction model, multi-scale architecture residual network for PPIs (MARPPI), based on dual-channel and multi-feature. Multi-feature leverages Res2vec to obtain the association information between residues, and utilizes pseudo amino acid composition, autocorrelation descriptors and multivariate mutual information to achieve the amino acid composition and order information, physicochemical properties and information entropy, respectively. Dual channel utilizes multi-scale architecture improved ResNet network which extracts protein sequence features to reduce protein feature loss. Compared with other advanced methods, MARPPI achieves 96.03%, 99.01% and 91.80% accuracy in the intraspecific datasets of Saccharomyces cerevisiae, Human and Helicobacter pylori, respectively. The accuracy on the two interspecific datasets of Human-Bacillus anthracis and Human-Yersinia pestis is 97.29%, and 95.30%, respectively. In addition, results on specific datasets of disease (neurodegenerative and metabolic disorders) demonstrate the ability to detect hidden interactions. To better illustrate the performance of MARPPI, evaluations on independent datasets and PPIs network suggest that MARPPI can be used to predict cross-species interactions. The above shows that MARPPI can be regarded as a concise, efficient and accurate tool for PPI datasets.