Search

As an effective representation of knowledge, knowledge graph network can be used to represent rich factual information between different categories and become an effective knowledge management tool. It has achieved great results in the application and research of knowledge engineering and artificial intelligence. Knowledge graph is usually expressed as a complex network structure. Its unstructured characteristics make the application of graph neural network to the analysis and research of knowledge graph become a research hotspot in academia. The purpose of this paper is to provide extensive research on knowledge graph construction technology based on graph neural network to solve two types of knowledge graph construction tasks, including knowledge extraction (entity, relationship and attribute extraction) and knowledge merging and processing (link prediction, entity alignment and knowledge reasoning, etc.). Through these tasks, the structure of knowledge graph can be further improved and new knowledge and reasoning relationships can be discovered. This paper also studies the advanced graph neural network method for knowledge graph related applications, such as recommendation system, question answering system and computer vision. Finally, the future research directions of knowledge graph application based on graph neural network are proposed. [ABSTRACT FROM AUTHOR]

Objective: To analyze RELT gene mutation found in a pedigree with clinical features and inheritable pattern consistent with amelogenesis imperfecta (AI) in China, and to study the relationship between its genotype and phenotype., Methods: Clinical and radiological features were recorded for the affected individuals. Peripheral venous blood samples of the patient and family members were collected for further study, and the genomic DNA was extracted to identify the pathogenic gene. Whole exome sequencing (WES) was performed to analyze the possible pathogenic genes, and Sanger sequencing was performed for validation. SIFT and PolyPhen-2 were used to predict and analyze the mutation effect. Comparison of RELT amino acids across different species were performed by using Uniprot website. In addition, the three-dimen-sional structures of the wild type and mutant proteins were predicted by Alphafold 2., Results: The proband exhibited typical hypocalcified AI, with heavy wear, soft enamel, rough and discolored surface, and partial enamel loss, while his parents didn ' t have similar manifestations. WES and Sanger sequencing results indicated that the proband carries a homozygous frameshift mutation in RELT gene, NM_032871.3: c.1169_1170del, and both of his parents were carriers. This mutation was predicted to be pathogenic by SIFT and PolyPhen-2. Up to now, there were 11 mutation sites in RELT gene were reported to be associated with AI, and all of the patients exhibited with hypocalcified AI. Compared with the wild-type RELT protein, the mutant protein p. Pro390fs35 conformation terminated prematurely, affecting the normal function of the protein., Conclusion: Through phenotype analysis, gene sequencing, and functional prediction of a Chinese family with typical amelogenesis imperfecta, this study found that RELT gene frameshift mutation can lead to protein dysfunction in AI patients. Further research will focus on the role and mechanism of RELT in enamel development at the molecular and animal levels, providing molecular biology evidence for the genetic counseling, prenatal diagnosis, and early prevention and treatment of AI.