Future potential and challenge of investigative toxicology with AI

DOI

抄録

<p>In recent years, innovative technologies beyond the conventional have begun to be developed and applied to drug discovery research. With innovations in iPS cell technology, gene editing and Omics-related technologies, the granularity of data has become finer than before, and it was realized that a huge amount of data can be handled in the safety evaluation process with multiple readouts. For example, according to a survey by a pharmaceutical company in the EU, high content analysis technology that can analyze the health status of cells from multiple angles is understanded as a game changer in several companies in 2015. It has been used as a practical tool by almost companies in 2020. On the other hand, retrospective analysis using biobanks not only contributes to the elucidation of disease mechanisms in human but is beginning to be applied to drug safety research. Using knowledge management databases integrating the results from complex searches with automated processing will help to handle the big data volumes. Thus, by leveraging knowledge management to accumulates various in vitro data and in vivo data and integrates clinical information, we may be able to predict drug safety more accurately in pre-clinical stage. Finally, machine learning and AI with the database could help to develop hypotheses of the toxicological mechanism for early candidates. In this presentation, I try to introduce examples of various in vitro data accumulation in our company and discuss the future potential and challenges for using AI in investigative toxicology.</p>

収録刊行物

詳細情報 詳細情報について

  • CRID
    1390580870561584000
  • DOI
    10.14869/toxpt.50.1.0_s7-2
  • 本文言語コード
    ja
  • データソース種別
    • JaLC
  • 抄録ライセンスフラグ
    使用不可

問題の指摘

ページトップへ