Construction of a multilayered integrated epigenetics database based on machine learning and deep learning

2020 Fiscal Year Final Research Report

• Project/Area Number: 18K11542
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 62010:Life, health and medical informatics-related
• Research Institution: National Cancer Center Japan
• Principal Investigator: Kaneko Syuzo 国立研究開発法人国立がん研究センター, 研究所, ユニット長 (10777006)
• Co-Investigator(Kenkyū-buntansha): 浜本 隆二 国立研究開発法人国立がん研究センター, 研究所, 分野長 (80321800)
• Project Period (FY): 2018-04-01 – 2021-03-31
• Keywords: データマイニング / ロボティクス / エピジェネティクス

Outline of Final Research Achievements

In this research project, we aimed to elucidate the true nature of cancer epigenetics at the individual level. Specifically, we used deep learning to classify cell image data in order to elucidate chromatin structure at the single cell level. In addition, we demonstrated that ChIP-seq analysis using robotics technology can be applied to various types of samples such as frozen samples and FFPE samples, and analyzed the obtained NGS data sets. As a result, we reported several peer-reviewed papers in English within this research period. In collaboration with the Artificial Intelligence Research Center of AIST, we implemented the FFPE ChIP-seq technology, which can also identify transcription factor binding sites, on a humanoid robot.

Free Research Field

情報科学、ロボティクス

Academic Significance and Societal Importance of the Research Achievements

本研究課題で得られた、FFPE検体を用いたChIP-seq解析手法は、ロボティクス技術を駆使したことによって大規模解析が可能となり、臨床学的に意義のあるデータを得ることが出来た。具体的には、FFPE検体に紐付けられている、予後・転移・薬剤感受性等の臨床情報と組み合わせることで、今後の創薬展開の道を拓いたと判断している。