Density Functional Theory in the Prediction of Mutagenicity: A Perspective

Abstract

As a field, computational toxicology is concerned with using in silico models to predict and understand the origins of toxicity. It is fast, relatively inexpensive, and avoids the ethical conundrum of using animals in scientific experimentation. In this perspective, we discuss the importance of computational models in toxicology, with a specific focus on the different model types that can be used in predictive toxicological approaches toward mutagenicity (SARs and QSARs). We then focus on how quantum chemical methods, such as density functional theory (DFT), have previously been used in the prediction of mutagenicity. It is then discussed how DFT allows for the development of new chemical descriptors that focus on capturing the steric and energetic effects that influence toxicological reactions. We hope to demonstrate the role that DFT plays in understanding the fundamental, intrinsic chemistry of toxicological reactions in predictive toxicology.

Characterization of applicability domains for QSAR models

“…Verhaar 分类框架 [83] 将化学品分为惰性、弱惰性、反应性、特异作用等类别 [83] [84] 。对特异性作用于受体蛋白的 终点，表征参与特异性作用的关键结构模式的碎片，以及分子在蛋白口袋的取向、 构象、诱导契合效应等，如结构警示 [85,86] 、CoMFA 及高维 QSAR 方法 [87,88] ，以及 借助分子对接、分子动力学模拟来计算构象系综的描述符 [89] (即分子起始事件，MIEs)为起点的关键事件序列 [90,91] 。此后，围绕 MIEs 的毒性测 试方法开发与验证的研究和技术导则显著增多 [92~94] 。例如，Tox21 项目 [57] 的测试 终点几乎全是 MIEs，很难再拆分为更基元的机理。尽管如此，QSAR 预测效果会 受到终点系统复杂性、空间异质性、或系统涌现 [78] 和混沌 [79] 的影响，对其微观机 制的思考总是裨益良多。 尤其在深度学习对化学分子的建模中，后者占据主导地位 [26] 。大数据和机器学习 时代的 QSAR 建模情景中，描述符往往具有较高的维度(≫3)，或者隐含在机器学 习流程内部，难以用经典的描述符域表征 [26,95] 。此时，本文介绍的基于分子指纹 相似度的结构域 [46,47] ，提供了一种可行的应用域定义方案。 仅采用小分子结构信息作为描述符，无法从复杂系统的空间异质性或涌现 [78] 的角度揭示机理，因而产生了活性悬崖 [54] 现象。此时，只有借助更多的数据提升 SAL 的分辨率，才有可能勾勒出活性悬崖边界，或识别出结构-活性关系较为连续 的化合物群。注意到，为应对不平衡或小样本数据问题，机器学习领域常基于相 似性原理构造人工样本，包括半监督学习或过采样方法 [96] 。然而，活性悬崖的存 在否定了相似性原理的绝对性。因此，需要谨慎使用构造人工样本的策略，以免 产生大量错误训练样本，而训练出错误的模型。 随着中国对化学品环境管理的加强，对新污染物治理的重视 of the "activity cliff" were further discussed and assumed to be dependent on the complexity and spatial heterogeneity of the investigated endpoint systems. Particularly, the chaos and emergence behavior of complex systems could be sensitive to tiny but specific structural changes in the small molecules and thus cannot be satisfactorily predicted based purely on the molecular descriptors.…”

In-silico investigation of oxoaporphine alkaloids of Xylopia aethiopica against SARS-COV-2 main protease

“…While isolation and characterization are critical steps in the identification of bioactive molecules in natural product drug discovery, computational techniques such as molecular docking, density functional theory (DFT), physiologically-based pharmacokinetic modelling, and others are becoming increasingly more popular for identifying of likely bioactive secondary metabolites in a timely and efficient manner [17]. For instance, the most widely used approach, DFT, has a lower computational cost than many other methods and is used to predict various properties of organic molecules including bond length, bond angles, dihedral angles and toxicity amongst others [18][19][20]. Molecular docking is commonly used to determine the appropriate orientation of molecules in the active site of the receptors of interest, and as well as determine their binding affinity.…”

Applications of Density Functional Theory on Heavy Metal Sensor and Hydrogen Evolution Reaction (HER)

“…This kind of databases will be the standard reference for the future research. In computational chemistry, the various DFT methods have been implemented and studied in several systems including chemosensor [22,23], hydrogen evolution reaction (HER) [24][25][26], oxygen reduction reaction (ORR) [27,28], oxygen evolution reaction (OER) [29], molecular machines [30,31], DNA mutation [32][33][34], selective etching [35,36], atomic layer deposition (ALD) [37,38] etc. Although a number of applications of DFT were reported, this book chapter precisely focused on heavy metal sensor and hydrogen evolution reaction.…”