Hierarchical Feature Extraction from Large Sequence Data by Deep Learning

2017 Fiscal Year Final Research Report

• Project/Area Number: 26330328
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Life / Health / Medical informatics
• Research Institution: Kanazawa University
• Principal Investigator: SATOU Kenji 金沢大学, 電子情報学系, 教授 (10215783)
• Project Period (FY): 2014-04-01 – 2018-03-31
• Keywords: テキスト分類 / 畳み込みニューラルネットワーク / スプライス部位 / プロモータ / 単語埋め込み / 次世代シーケンサ / ゲノム配列決定

Outline of Final Research Achievements

In the field of biological sequence analysis including DNA and amino acid sequence analysis, traditional methods are highly dependent on the knowledge specific to molecular biology, so their ability is limited to the discovery of features easily predicted from domain-specific knowledge. In this study, it is shown that by using various machine learning algorithms including deep learning, it is possible to extract novel and hierarchical features from large sequence data.

Free Research Field

生命情報学