• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2022 Fiscal Year Final Research Report

Risk prediction modeling by accounting for interaction between health-related data and whole-genome information

Research Project

  • PDF
Project/Area Number 20K11723
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 60030:Statistical science-related
Research InstitutionNagasaki University

Principal Investigator

Ueki Masao  長崎大学, 情報データ科学部, 教授 (10515860)

Project Period (FY) 2020-04-01 – 2023-03-31
Keywords予測モデル / 遺伝子x環境相互作用 / 高次元回帰モデル / 健康医療データ / ゲノムデータ
Outline of Final Research Achievements

Combining whole-genome data with various health-related data, we developed statistical models and algorithms for risk prediction. In particular, we developed prediction model that incorporates interactions between genome-data and health-related data. Based on the STMGP (smooth-threshold multivariate genetic prediction), a sparse modeling method, we evaluated the prediction model combing whole-genome data and other factors such as sex and age on real dataset. Subsequently, we developed a prediction model that incorporates whole-genome data, non-genomic data (sex, age, etc), and their interactions, which is a non-additive gene-environment interaction based prediction model.

Free Research Field

統計学

Academic Significance and Societal Importance of the Research Achievements

近年、ゲノムデータを含め、高次元な健康医療データが取得されているが、十分な疾患リスク予測精度が得られないケースが多くある。本研究において、現行の単純な加法モデルを発展させることで、全ゲノム情報と多様な健康医療データの相互作用を考慮できる非線形リスク予測モデルを開発した。これまでゲノムデータに対する予測モデルにおいて非ゲノムデータとの相互作用を考慮できる予測モデルは限られていたが、本手法を用いることで、ゲノムデータと健康医療データの相互作用が存在する場合の予測精度向上に貢献するものと考える。

URL: 

Published: 2024-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi