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

Modeling for Prediction of Serious Adverse Events Probabilities of Drug Candidates

Research Project

Project/Area Number 15KT0017
Research Category

Grant-in-Aid for Scientific Research (B)

Allocation TypeMulti-year Fund
Section特設分野
Research Field Mathematical Sciences in Search of New Cooperation
Research InstitutionOsaka University

Principal Investigator

TAKAGI Tatsuya  大阪大学, 薬学研究科, 教授 (80144517)

Co-Investigator(Kenkyū-buntansha) 日比 孝之  大阪大学, 情報科学研究科, 教授 (80181113)
岡本 晃典  北陸大学, 薬学部, 講師 (70437309)
川下 理日人  大阪大学, 薬学研究科, 助教 (00423111)
田 雨時  大阪大学, 薬学研究科, 助教 (60761252)
Project Period (FY) 2015-07-10 – 2020-03-31
Project Status Completed (Fiscal Year 2019)
Budget Amount *help
¥14,560,000 (Direct Cost: ¥11,200,000、Indirect Cost: ¥3,360,000)
Fiscal Year 2018: ¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000)
Fiscal Year 2017: ¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000)
Fiscal Year 2016: ¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000)
Fiscal Year 2015: ¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000)
Keywords有害事象 / 機械学習 / ATCコード / Stacking / 化学記述子 / 悪性症候群 / スティーブンス・ジョンソン症候群 / マルチラベル学習 / 重篤希少有害事象 / AI / ROC曲線 / 多変量解析 / 副作用予測 / サポートベクターマシン / ランダムフォレスト / 多数決法 / 重大副作用 / JADER / Random Forest / 医薬品有害事象 / グレブナー基底
Outline of Final Research Achievements

We tried to make some models for predicting drugs which show relatively high probabilities of rare and severe adverse events using chemical structure information and machine learning methods. As results, several machine learning methods (Logistic Regression, Random Forest, Support Vector Machine, Artificial Neural Network, etc) with Stacking method showed enough prediction abilities when ATC code was introduced for malignant syndrome and thrombocytopenia. These results can be utilized as a powerful tool for Drug Development and drug administration guidance.

Academic Significance and Societal Importance of the Research Achievements

希少有害事象はこれまで、ともすれば、見過ごされがちであった。理由としては、極めて珍しいこと、殆どの医薬品に見られるので、これを咎めると、医薬品が市場に無くなってしまうことが考えられる。しかしながら中には命に関わる重篤なものもあり、かつ、一部の医薬品で高確率で見られ、回収に繋がったものもある事から、このまま見過ごすことはできないと考えた。今回の結果により、より高確率で希少重篤有害事象を引き起こす医薬品の化学的特徴が明らかとなり、将来的には、医薬品開発の様々な段階で援用され、服薬指導時に注意を促すための材料となり、有害事象から命を落としたり、後遺症に悩む人々の現出を予防できることが期待される。

Report

(6 results)
  • 2019 Annual Research Report   Final Research Report ( PDF )
  • 2018 Research-status Report
  • 2017 Research-status Report
  • 2016 Research-status Report
  • 2015 Research-status Report
  • Research Products

    (5 results)

All 2019 2018 2017 2016

All Presentation (5 results) (of which Int'l Joint Research: 2 results,  Invited: 1 results)

  • [Presentation] 機械学習を用いた悪性症候群予測モデルの構築2019

    • Author(s)
      望月麻衣、福戸康平、田雨時、高木達也
    • Organizer
      第47回構造活性相関シンポジウム(熊本)
    • Related Report
      2019 Annual Research Report
  • [Presentation] 機械学習を用いた医薬品の有害事象の予測モデル構築の検討2018

    • Author(s)
      Ni Tao、高木達也、日比孝之、望月麻衣、森脇寛智、田雨時
    • Organizer
      第46回構造活性相関シンポジウム
    • Related Report
      2018 Research-status Report
  • [Presentation] Prediction of Serious Adverse Events Using Machine Learning2018

    • Author(s)
      Yushi-Tian, Hirotomo Moriwaki, Hiroaki Moriuchi, Satoshi Aoki, Nobuki Takayama, Norihito Kawashita, Takayuki Hibi, Tatsuya Takagi
    • Organizer
      19th International Conference on Medicinal Chemistry and Multi Targeted Drug Delivery
    • Related Report
      2018 Research-status Report
    • Int'l Joint Research / Invited
  • [Presentation] Prediction of Serious Adverse Events Using Machine Learning2017

    • Author(s)
      Tatsuya TAKAGI
    • Organizer
      11th Asian Federation for Medicinal Chemistry's International Medicinal Chemistry Symposium
    • Related Report
      2017 Research-status Report
    • Int'l Joint Research
  • [Presentation] Constructing prediction models of adverse drug reactions using machine learning2016

    • Author(s)
      Hiroaki MORIUCHI
    • Organizer
      44th Symposium on Structure-Activity Relationships
    • Place of Presentation
      Kyoto
    • Year and Date
      2016-11-16
    • Related Report
      2016 Research-status Report

URL: 

Published: 2015-07-14   Modified: 2022-11-04  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi