Statistical methodology for stabilizing of the genome-omics data analysis

2017 Fiscal Year Final Research Report

• Project/Area Number: 25280008
• 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 Institute of Statistical Mathematics
• Principal Investigator: Egichi Shinto 統計数理研究所, 数理・推論研究系, 教授 (10168776)
• Co-Investigator(Kenkyū-buntansha): 松浦 正明 帝京大学, 公私立大学の部局等, 教授 (40173794) ; 松井 茂之 名古屋大学, 医学系研究科, 教授 (80305854) ; 小森 理 福井大学, 学術研究院工学系部門, 講師 (60586379) ; 間野 修平 統計数理研究所, 数理・推論研究系, 准教授 (20372948) ; 野間 久史 統計数理研究所, データ科学研究系, 准教授 (70633486)
• Co-Investigator(Renkei-kenkyūsha): TAKENOUCHI Takashi 公立はこだて未来大学, システム情報科学部, 准教授 (50403340) ; HENMI Masayuki 統計数理研究所, データ科学研究系, 准教授 (80465921)
• Research Collaborator: COPAS John B. University of Warwick, Department of Statistics, Emeritus Professor ; Huang Su-Yun Academia Sinica, Institute of Statistical Science ; HUNG HUNG National Taiwan University, Graduate Institute of Epidemiology and Preventive Medicine, Associate Professor
• Project Period (FY): 2013-04-01 – 2018-03-31
• Keywords: オミクス データ / 遺伝子ランキング / データ異質性 / 高次元・小標本 / 準線形モデル / 表現型予測

Outline of Final Research Achievements

In genome and omics data analysis we aimed at exploiting new statistical methods to challenge a difficult aspect caused by high-dimensional small-sample observations. In particular, we have proposed and published statistical methods and practical algorithms for prediction for treatment effect and prognosis based on genome and omics data beyond existing methods. In major contributions the subsampling replication method for a gene ranking was presented for a stable ranking, and a new modeling, called quasi-linear model, was proposed to incorporate different linear predictors into an integrated predictor connecting by a generalized mean.

Free Research Field

統計科学