Department of Chemistry , University of Bergen , Allégaten 41 , N-5007 Bergen , Norway.
Department of Chemistry , Norwegian University of Science and Technology , N-7491 Trondheim , Norway.
J Chem Inf Model. 2019 Oct 28;59(10):4077-4082. doi: 10.1021/acs.jcim.9b00516. Epub 2019 Sep 18.
A general-purpose software package, termed DE Novo OPTimization of In/organic Molecules (DENOPTIM), for design and virtual screening of functional molecules is described. Molecules of any element and kind, including metastable species and transition states, are handled as chemical objects that go beyond valence-rules representations. Synthetic accessibility of the generated molecules is ensured via detailed control of the kinds of bonds that are allowed to form in the automated molecular building process. DENOPTIM contains a combinatorial explorer for screening and a genetic algorithm for global optimization of user-defined properties. Estimates of these properties may be obtained to form the fitness function (figure of merit or scoring function) from external molecular modeling programs via shell scripts. Examples of a range of different fitness functions and DENOPTIM applications, including an easy-to-do test case, are described. DENOPTIM is available as Open Source from https://github.com/denoptim-project/DENOPTIM .
描述了一种通用软件包，称为用于设计和虚拟筛选功能分子的有机/无机分子的从头优化（DENOPTIM）。可以将任何元素和种类的分子，包括亚稳物种和过渡态，作为化学对象进行处理，这些对象超越了价规则表示。通过对允许在自动分子构建过程中形成的键的种类进行详细控制，确保生成的分子具有合成可及性。DENOPTIM 包含用于筛选的组合探索器和用于用户定义属性的全局优化的遗传算法。可以从外部分子建模程序通过外壳脚本获得这些属性的估计值，以形成适合度函数（优值函数或评分函数）。描述了一系列不同的适合度函数和 DENOPTIM 应用程序的示例，包括一个易于执行的测试用例。DENOPTIM 可从 https://github.com/denoptim-project/DENOPTIM 获得开源。
