文獻探勘系統於尋找疾病的相關基因之應用 Text-mining System for Disease Associated Gene http://bws.iis.sinica.edu.tw/THOD/ 目標 Object Researchers are finding it more and more difficult to follow the changing status of disease candidate genes due to the exponential increase in gene mapping studies. The Text-mined Hypertension, Obesity, and Diabetes candidate gene database (THOD), are developed to help trace existing research on three complex diseases (hypertension, obesity, and diabetes), by regularly and semi-automatically extracting HOD-related genes from new published literature. T-HOD employed state-of-art text-mining technologies including a named entity recognition-gene normalization (GN) system and a disease-gene relation extraction system. Unlike manually constructed disease gene databases, such as the genetic association database (GAD), the content of T-HOD is regularly updated by our textmining system and verified by domain experts. 系統 System The flowchart of T-HOD database construction. 使用介面 Interface User interface of T-HOD database. We divided T-HOD user interface into four regions for introducing function of interface clearly. The Network viewer shows the network viewer that presents a graphic-based gene-gene network for a selected candidate gene. The statistics of the extracted hypertension candidate genes. The blue bars indicate the number of genes extracted each year, while the yellow bars show the number of new discovered genes each year. 運用資料探勘技術從事廣泛的基因篩選及 endophenotype 構建 在高血壓疾病上之應用 計畫主持人 許聞廉 研究員 智慧型代理人系統實驗室 Intelligent Agent Systems Lab