Development of a graphical model and a tolerance region for the assessment of physiological process and ecological diversity

2015 Fiscal Year Final Research Report

• Project/Area Number: 25280006
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Partial Multi-year Fund
• Section: 一般
• Research Field: Statistical science
• Research Institution: The University of Tokyo
• Principal Investigator: Kishino Hirohisa 東京大学, 農学生命科学研究科, 教授 (00141987)
• Co-Investigator(Kenkyū-buntansha): Kitada Shuichi 東京海洋大学, 海洋科学技術研究科, 教授 (10262338)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 医薬生物・ゲノム解析 / 生体と生態系の健全性 / 遺伝子ネットワーク / 分子系統樹 / リスク評価

Outline of Final Research Achievements

First, we developed a method that constructs the entropy-based maximal core connected subgraph of gene expression profiles. The analysis of the expression profiles at the early developmental stage of Atlantic salmon showed that the resultant graph makes it easy to understand the key players and their interactions. Second, we defined two complementary tolerance intervals. The coverage probability of the standard tolerance interval is at least the pre-specified level, whereas that of the complementary tolerance interval is at most the specified level. With the two complementary tolerance intervals, we developed a statistical procedure to test a target population includes a pre-specified amount of outliers with respect to the reference population. Third, we developed a new index of community diversity. By comparing the species sampling proportion with the value estimated assuming random sampling, it measures phylogenetic skew of member species.

Free Research Field

統計科学、分子進化学