Sui MingShuang, Cui Lei
School of Medical Informatics, China Medical University, Shenyang, Liaoning, China.
Stud Health Technol Inform. 2017;245:531-535.
Our objective was to identify and extract gene-drug and drug-adverse drug reaction (ADR) relationships from different biomedical literature collections, and to predict the possible association between gene and ADR. The drug, ADR and gene entities were recognized by a CRF model with multiple features. Logistic regression models were constructed for each drug-ADR and drug-gene pair based on its frequency, Mesh Rule association and similarity with known association etc. Using predicted score to generate drug-ADR matrix and drug-gene matrix, and then calculating for gene-ADR matrix. Network and clustering analysis were applied to verify and interpret the relationship between them. A total of 78014 potential gene-ADR associations were predicted. Part of the predicted results can be explained by the network-clustering-pathway analysis, and verified in the literature. The gene-drug-ADR network constructed in this study can provide a reference for the possible association between the gene and ADR.
我们的目标是从不同的生物医学文献集中识别并提取基因-药物和药物-药物不良反应（ADR）关系，并预测基因与ADR之间的可能关联。药物、ADR和基因实体通过具有多个特征的CRF模型来识别。基于每种药物-ADR和药物-基因对的频率、医学主题词（Mesh）规则关联以及与已知关联的相似性等构建逻辑回归模型。利用预测分数生成药物-ADR矩阵和药物-基因矩阵，然后计算基因-ADR矩阵。应用网络和聚类分析来验证和解释它们之间的关系。共预测出78014个潜在的基因-ADR关联。部分预测结果可通过网络-聚类-通路分析得到解释，并在文献中得到验证。本研究构建的基因-药物-ADR网络可为基因与ADR之间的可能关联提供参考。
