NGS analysis of a large genome cohort by deep learning

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K06625
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 43060:System genome science-related
• Research Institution: Tohoku University
• Principal Investigator: Takayama Jun 東北大学, 未来型医療創成センター, 准教授 (20574114)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: ヒトゲノム解析 / 次世代シークエンシング

Outline of Final Research Achievements

This study is intended to identify the technical limitations of NGS, one of the major genome analysis techniques for large-scale genome cohort and disease analysis, and to overcome these limitations using state-of-the-art technologies such as deep learning. Under this study, we focused on the bias due to population differences among the biases latent in the resequencing method, a genome analysis method using NGS, and evaluated the performance of the reference genome sequence JG1, which is optimized for the Japanese population to replace the international reference genome constructed in the Human Genome Project. The results, together with other results, were published in Nature Communications. I also gave several invited lectures.

Free Research Field

ゲノム科学

Academic Significance and Societal Importance of the Research Achievements

次世代シークエンシング法を用いるとヒトの全ゲノム情報を数日の内に解読可能であるが、完璧な方法ではない。本研究では次世代シークエンシング法の問題点を追求し、特に、民族集団の違いによるバイアスを克服することを試みた。より具体的には日本人のゲノム解析に最適化した日本人基準ゲノム配列JG1の性能評価を行った。日本人基準ゲノム配列は公開され、ゲノム解析サービスに利用されており、本研究はその有用性を示すものとして大きな意義を有すると考えられる。