摘要：本文基于通用数据资源(科技文献、科研活动等)和专业领域知识资源(如组学科研数据),以水稻粒型基因领域为例,探索具有一定普适性,能兼顾知识覆盖广度和深度(全景式),并可充分继承整合多源异构数据和知识(多路径)的领域知识图谱构建方法。首先,继承复用权威学术论文中专家先验知识和多种领域本体,自顶向下设计构建全景式水稻粒型基因知识图谱模式层的本体模型;其次,通过图数据抽取、结构化及半结构化转换映射和非结构化文本抽取等多路径实现图谱数据实例填充,并基于数据挖掘发现的新实体及其语义关系,进行自底向上的本体模型迭代完善;再次,通过实体消歧、实体链接等实现多源知识关联融合,并基于Neo4j数据库实现图谱数据持久化存储;最后,对领域知识图谱驱动下的典型知识关联与发现服务应用场景进行展望。实验结果表明,本文研究提出的全景式、多路径领域知识图谱构建方法,具有一定集成性和通用性,可为细分垂直领域的知识图谱构建提供参考。

摘要：Extremely small-sized iron oxide nanoparticles(ESIONPs)with sizes less than 5 nm have shown great promise as T_(1)contrast agents for magnetic resonance imaging(MRI).However,their facile and scalable production with simultaneously endowed biocompatible surface chemistry remains difficult to be *** this study,by using the coprecipitation method implemented in a specially designed gas/liquid mixed phase fluidic reactor,polyglucose sorbitol carboxymethyether(PSC)coated ESIONPs were continuously synthesized with controllable particle sizes ranging from 1.8 to 4 *** the differently sized ESIONPs,the 3.7-nm ESIONPs exhibit the best performance as T_(1)MRI contrast agent,featuring a high r_(1) value of 4.11(mmol L^(−1))^(−1)s^(−1)and low r_(2)/r_(1) ratio of 7.90 under a clinical 3 T MR scanning,as well as the excellent T_(1)MRI contrast effect in not only water but also the cellular environment and blood ***,the ESIONPs possess long-term stability and good dispersity in aqueous dispersions,making them ideal candidates as safe and effective T_(1)-weighted MRI contrast agent for real clinical use.