Crop trait regulating-genes knowledge graph dataset

In the scientific research of crop breeding, breeding new crop varieties with various excellent traits has always been the direction of efforts of breeders. At present, with the accelerated application of information technology in the field of crop breeding, the multi-dimensional scientific data related to crop breeding has shown exponential growth. These semi-structured and structured scientific data are distributed in scientific databases in different fields and lack the association and fusion of multi-dimensional scientific data across species. It hindered the transfer and reuse of existing crop breeding knowledge and maximized the value of crop breeding scientific data, which brought challenges to the knowledge discovery of crop trait regulation genes. Therefore, more and more crop breeding research work is based on the reorganization, correlation, analysis and utilization of existing breeding scientific data, so as to achieve the discovery of crop trait regulation gene knowledge.The dataset of knowledge map of crop trait regulatory genes was selected from PubMed literature database, Phytozome (genomic information of 4 species) and Ensembl (European Molecular Biology Laboratory's European) Bioinformatics Institute (Bioinformatics Institute) plants (Genome information of 4 species), UniProt (Universal Protein) (protein Annotation information of 4 species), Rice Genome Annotation (RGAP) Project), STRING (protein interaction information for 4 species), Pfam (Protein family analysis and modeling) (protein family information for 4 species), KEGG (Kyoto Encyclopedia of Genes) The entities and relationships of the multi-source scientific data with different data formats were extracted using the and Genomes (pathway annotation information of the 4 species) and the GO (Gene Ontology) domain scientific database as the data sources. It mainly includes mapping knowledge extraction for structured data. For XML semi-structured data, knowledge extraction based on Kettle data analysis is adopted. For FASTA semi-structured data, knowledge extraction based on BLAST model is adopted. For Text unstructured data, knowledge extraction based on large language model is adopted. On the basis of the above entity and relationship extraction, the association fusion of multi-source crop breeding knowledge was realized based on entity mapping and specific attribute association. Finally, the crop trait regulatory gene knowledge map dataset was formed, which consisted of 13 entity datasets and 16 entity relationship datasets.The crop trait -egulating gene knowledge graph dataset provides a key semantic model and important data basis for crop breeding knowledge discovery, such as excellent pleiotropic gene discovery, cross-species gene function prediction and potential discovery of pathway gene network.