现代计算机2024,Vol.30Issue(2) :66-71.DOI:10.3969/j.issn.1007-1423.2024.02.011

基于知识图谱的马表型-基因知识库构建与应用研究

Construction and application research of equine Phenotype-Gene knowledge database based on knowledge graph

郭迎春 丰建海 郭迎凤
现代计算机2024,Vol.30Issue(2) :66-71.DOI:10.3969/j.issn.1007-1423.2024.02.011

基于知识图谱的马表型-基因知识库构建与应用研究

Construction and application research of equine Phenotype-Gene knowledge database based on knowledge graph

郭迎春 1丰建海 2郭迎凤3
扫码查看

作者信息

  • 1. 内蒙古农业大学计算机与信息工程学院,呼和浩特 010018;内蒙古自治区农牧业大数据研究与应用重点实验室,呼和浩特 010018
  • 2. 内蒙古蒙牛乳业(集团)股份有限公司数字科创部,呼和浩特 011500
  • 3. 内蒙古蒙牛乳业(集团)股份有限公司奶源事业部,呼和浩特 011500
  • 折叠

摘要

马表型-基因知识库为马的遗传育种和疾病诊治提供辅助参考.从生物医学文献库Pubmed中提取相关文献摘要,采用多实体识别接口Pubtator进行生物实体识别,以半自动化方式通过公共域关系抽取工具OpenIE和人工标注相结合的方法实现马表型-基因知识图谱的构建.知识图谱包含了马的25种常见表型,分析获取到与之关联的基因、变异等实体139个,语义关系177个.马表型-基因知识图谱的构建可以将马科研工作者从繁琐耗时的文献检索中解脱出来,为进一步的研究提供便利,同时也为构建完整的马知识图谱提供技术参考.

Abstract

The equine phenotype-gene database provides auxiliary reference for genetic breeding and disease diagnosis and treatment of equines.The abstracts of related articles were extracted from biomedical literature database Pubmed,and the multi-entity recognition interface Pubtator was used for biological entity recognition.The equine gene-phenotype knowledge graph was constructed in a semi-automated manner by combining the public domain relationship extraction tool OpenIE and artificial annota-tion.The knowledge graph included 25 common phenotypes of equine,and 139 entities such as genes and variations associated with them and 177 semantic relationships were obtained.The Equine Phenotype-Gene Knowledge Graph frees equine researchers from tedious and time-consuming literature searches and facilitates further research,as well as providing a technical reference for constructing a complete equine knowledge graph.

关键词

知识图谱/文献挖掘/命名实体识别/关系抽取/马表型

Key words

knowledge graph/literature mining/named entity identification/relationship extraction/horse phenotype

引用本文复制引用

出版年

2024
现代计算机
中大控股

现代计算机

影响因子:0.292
ISSN:1007-1423
参考文献量15
段落导航相关论文