化学構造からの有害性発現予測：人工知能技術の適用

DOI Web Site Web Site PubMed 参考文献13件

書誌事項

タイトル別名
  • AI-based QSAR Modeling for Prediction of Active Compounds in MIE/AOP
  • Symposium Review 化学構造からの有害性発現予測 : 人工知能技術の適用
  • Symposium Review カガク コウゾウ カラ ノ ユウガイセイ ハツゲン ヨソク : ジンコウ チノウ ギジュツ ノ テキヨウ

この論文をさがす

抄録

<p>Toxicity testing is critical for new drug and chemical development process. A clinical study, experimental animal models, and in vitro study are performed to evaluate the safety of a new drug. The limitations of these methods include extensive time for toxicity testing, an ethical problem, and high costs of experimentation. Therefore computational methods are considered useful for estimating chemical toxicity. In silico toxicity prediction is one of the toxicity assessments that uses computational methods to predict and stimulate the toxicity of chemicals. In silico study aims to contribute to effective development of new drug and chemical design. In this study, quantitative structure-activity relationship (QSAR) models will be used to predict toxicities based on chemical structural parameters. Because toxicities are complicated physiological phenomena, a similar toxicity expression might cause a different pathway. Also, since many drugs with unknown mechanisms of actions are available, the application of artificial intelligence (AI)—which uses sophisticated algorithms— is increasingly used to predict toxicities. Recently, the QSAR model was applied to determine complex relations between chemical structures and toxicities. However, accuracy of QSAR for toxicity prediction remains an important issue. International competitions funded by public institutions can address this issue. Two important toxicity challenges were organized in the past decade; this article presents issues of toxicity based on these challenges.</p>

収録刊行物

  • 薬学雑誌

    薬学雑誌 140 (4), 499-505, 2020-04-01

    公益社団法人 日本薬学会

参考文献 (13)*注記

もっと見る

キーワード

詳細情報 詳細情報について

問題の指摘

ページトップへ