Abstract

본 연구에서는 췌장암의 유전자-단백질 상호작용 네트워크를 구성하고, 관련 연구에서 주요하게 언급되는 유전자-단백질의 유발관계 사슬을 파악함으로써, 췌장암의 원인을 규명하는 실증적인 연구로 이어질 수 있는 미발견 공공 지식을 제공하려 하였다. 이를 위하여 텍스트마이닝과 주경로 분석을 Swanson의 ABC 모델에 적용해 중간 개념인 B를 방향성을 가진 다단계 모델로 확장하고 가장 의미 있는 경로를 도출하였다. 본 연구의 주제가 된 췌장암의 사례처럼 시작점과 끝점조차 한정할 수 없는 미발견 공공 지식 추론에서 주경로 분석은 유용한 도구가 될 수 있을 것이다. This study aims to infer the gene-protein 'brings_about' chains of pancreatic cancer which were referred to in the pancreatic cancer related researches by constructing the gene-protein interaction network of pancreatic cancer. The chains can help us uncover publicly unknown knowledge that would develop as empirical studies for investigating the cause of pancreatic cancer. In this study, we applied a novel approach that grafts text mining and the main path analysis into Swanson's ABC model for expanding intermediate concepts to multi-levels and extracting the most significant path. We carried out text mining analysis on the full texts of the pancreatic cancer research papers published during the last ten-year period and extracted the gene-protein entities and relations. The 'brings_about' network was established with bio relations represented by bio verbs. We also applied main path analysis to the network. We found the main direct 'brings_about' path of pancreatic cancer which includes 14 nodes and 13 arcs. 9 arcs were confirmed as the actual relations emerged on the related researches while the other 4 arcs were arisen in the network transformation process for main path analysis. We believe that our approach to combining text mining analysis with main path analysis can be a useful tool for inferring undiscovered knowledge in the situation where either a starting or an ending point is unknown.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call