生成式深度学习在目标导向分子设计中的应用进展-中国材料进展

文章信息/Info

Title:: Application of Generative Deep Learning in Object-oriented Molecular Design

Author(s):: WANG Jifeng; WANG Ying; State Key Laboratory of Molecular Engineering of Polymers, Department of Macromolecular Science, Fudan University, Shanghai 200438, China

Keywords:: Molecule design; generative deep learning; generative adversarial network; variational autoencoder; denoising diffusion probabilistic model; model evaluation framework; molecular representation

摘要:: 分子设计作为化学与材料科学中的一项核心任务，面临着在庞大的化学空间中高效筛选并开发具备特定功能分子的问题，传统方法在效率和探索性方面存在明显局限。近年来，生成式深度学习的兴起为分子设计提供了自动化与智能化的新路径。本文综述了生成式深度学习在分子设计中的应用进展，首先对不同分子表示方法（如SMILES、分子图和三维结构表示）进行比较，分析了各自的优缺点。随后，综合评估了三种主流生成式模型：生成对抗网络（GAN）、变分自动编码器（VAE）和去噪扩散概率模型（DDPM），并探讨了生成式模型在特性控制、泛化能力和合成可行性等方面的评估方式。最后，本文聚焦于以目标为导向的分子设计，分析了生成式模型在分子生成质量、特性优化等方面的优势与不足，并基于现有技术的进展，提出了未来生成式模型在分子设计领域的研究方向。

Abstract:: Designing molecules with specific functions within an immense chemical space is a fundamental challenge in chemistry and materials science, as traditional methods often lack efficiency and exploratory capacity. The advent of generative deep learning has introduced automated and intelligent approaches that promise to transform molecular design. In this review, we examine the advancements in applying generative deep learning to molecular design. We first compare molecular representation methods, including SMILES notation, molecular graphs, and three-dimensional structural representations, highlighting their respective advantages and limitations. We then critically evaluate three leading generative models: Generative Adversarial Network (GAN), Variational Autoencoder (VAE), and Denoising Diffusion Probabilistic Model (DDPM), focusing on their capabilities in property control, generalization ability, and synthetic feasibility. We also discuss the characteristics of the three models in object-oriented molecular design by analyzing their capability of molecular generation and property optimization. Finally, we propose future research directions for leveraging the generative models in molecular design, aiming to inspire further advancements in this rapidly evolving field.