人工智能驱动的虚拟筛选在药物发现中的研究进展与应用

Research Progress and Application of Virtual Screening Driven by Artificial Intelligence in Drug Discovery

关键词:  虚拟筛选  大数据  CADD  AIDD

Key words:  virtual screening  big data  CADD  AIDD

摘要:

随着药物发现领域数据的快速增长和人工智能技术(Artificial Intelligence, AI)的发展,计算机辅助药物设计方法(Computer aided drug design,CADD)和人工智能辅助药物设计(Artificial Intelligence-driven Drug Design,AIDD)在药物筛选中的应用日益广泛。本文综述了虚拟筛选技术和人工智能(AI)在药物发现领域的应用,特别关注了基于配体的虚拟筛选(LBVS)、基于结构的虚拟筛选(SBVS)以及AI驱动的虚拟筛选技术。通过深入分析这些技术的优势与局限,文章指出AI技术特别是机器学习和深度学习算法的引入,极大地提升了虚拟筛选的准确性和效率,同时也面临数据质量、模型复杂性、泛化能力以及可解释性等挑战。未来研究应侧重于提升模型性能,同时优化其泛化能力和透明度,确保技术在实际应用中的有效性和可靠性。此外,深度生成模型等AI技术在探索药物化学空间方面展示出突破传统界限的潜力,预示着药物发现和分子设计领域的革命性变化。

Abstract:

With the rapid growth of data in drug discovery, the application of Artificial Intelligence (AI) in drug screening is becoming increasingly widespread. This article reviews the application of virtual screening technology and Artificial Intelligence (AI) in drug discovery, with a particular focus on ligand-based virtual screening (LBVS), structure-based virtual screening (SBVS), and AI-driven virtual screening technologies. Through an in-depth analysis of the strengths and limitations of these technologies, the article points out that AI technologies, especially the introduction of machine learning and deep learning algorithms, have greatly improved the accuracy and efficiency of virtual screening, while also facing challenges such as data quality, model complexity, generalization ability, and interpretability. Future research should focus on improving model performance while optimizing its generalization ability and transparency to ensure the effectiveness and reliability of the technology in practical applications. Additionally, AI technologies such as deep generative models demonstrate the potential to break traditional boundaries in exploring the chemical space of drug discovery, heralding revolutionary changes in the field of drug discovery and molecular design.