Enhancing Cross-Lingual Entity Alignment in Knowledge Graphs through Structure Similarity Rearrangement.

Cross-lingual entity alignment in knowledge graphs is a crucial task in knowledge fusion. This task involves learning low-dimensional embeddings for nodes in different knowledge graphs and identifying equivalent entities across them by measuring the distances between their representation vectors. Existing alignment models use neural network modules and the nearest neighbors algorithm to find suitable entity pairs. However, these models often ignore the importance of local structural features of entities during the alignment stage, which may lead to reduced matching accuracy. Specifically, nodes that are poorly represented may not benefit from their surrounding context. In this article, we propose a novel alignment model called SSR, which leverages the node embedding algorithm in graphs to select candidate entities and then rearranges them by local structural similarity in the source and target knowledge graphs. Our approach improves the performance of existing approaches and is compatible with them. We demonstrate the effectiveness of our approach on the DBP15k dataset, showing that it outperforms existing methods while requiring less time.

Semantic Health Knowledge Graph: Semantic Integration of Heterogeneous Medical Knowledge and Services.

With the explosion of healthcare information, there has been a tremendous amount of heterogeneous textual medical knowledge (TMK), which plays an essential role in healthcare information systems. Existing works for integrating and utilizing the TMK mainly focus on straightforward connections establishment and pay less attention to make computers interpret and retrieve knowledge correctly and quickly. In this paper, we explore a novel model to organize and integrate the TMK into conceptual graphs. We then employ a framework to automatically retrieve knowledge in knowledge graphs with a high precision. In order to perform reasonable inference on knowledge graphs, we propose a contextual inference pruning algorithm to achieve efficient chain inference. Our algorithm achieves a better inference result with precision and recall of 92% and 96%, respectively, which can avoid most of the meaningless inferences. In addition, we implement two prototypes and provide services, and the results show our approach is practical and effective.

Characterization of a Novel Nicotine Degradation Gene Cluster ndp in Sphingomonas melonis TY and Its Evolutionary Analysis.

Sphingomonas melonis TY utilizes nicotine as a sole source of carbon, nitrogen, and energy through a variant of the pyridine and pyrrolidine pathways (VPP). A 31-kb novel nicotine-degrading gene cluster, ndp, in strain TY exhibited a different genetic organization with the vpp cluster in strains Ochrobactrum rhizosphaerae SJY1 and Agrobacterium tumefaciens S33. Genes in vpp were separated by a 20-kb interval sequence, while genes in ndp were localized together. Half of the homolog genes were in different locus in ndp and vpp. Moreover, there was a gene encoding putative transporter of nicotine or other critical metabolite in ndp. Among the putative nicotine-degrading related genes, the nicotine hydroxylase, 6-hydroxy-L-nicotine oxidase, 6-hydroxypseudooxynicotine oxidase, and 6-hydroxy-3-succinyl-pyridine monooxygenase responsible for catalyzing the transformation of nicotine to 2, 5-dihydropyridine in the initial four steps of the VPP were characterized. Hydroxylation at C6 of the pyridine ring and dehydrogenation at the C2-C3 bond of the pyrrolidine ring were the key common reactions in the VPP, pyrrolidine and pyridine pathways. Besides, VPP and pyrrolidine pathway shared the same latter part of metabolic pathway. After analysis of metabolic genes in the pyridine, pyrrolidine, and VPP pathways, we found that both the evolutionary features and metabolic mechanisms of the VPP were more similar to the pyrrolidine pathway. The linked ndpHFEG genes shared by the VPP and pyrrolidine pathways indicated that these two pathways might share the same origin, but variants were observed in some bacteria. And we speculated that the pyridine pathway was distributed in Gram-positive bacteria and the VPP and pyrrolidine pathways were distributed in Gram-negative bacteria by using comprehensive homologs searching and phylogenetic tree construction.