Data underlying the chapter: Machine-Learning-Guided Optimization of Phosphine-based Ligands for Nickel-Catalyzed Addition of Arylboronic Acids to Nitriles

We used high-throughput experimentation, density functional theory and machine learning to guide optimization of bisphosphine ligands for the nickel-catalyzed addition of arylboronic acids to nitriles. This dataset contains the version of the supporting information as published with this chapter, all code and data to reproduce the results and use the same approach on new datasets, an overview of the calculated descriptors, an overview of the ligands and the experimental results and finally an interactive version of the ensemble prediction made with the transfer learning approach presented in this paper.

本研究采用高通量实验、密度泛函理论（density functional theory）与机器学习方法，对镍催化芳基硼酸与腈的加成反应所用双膦配体开展优化导向研究。本数据集包含与本章节同步发表的支持信息版本、可复现本研究结果并将该方法推广至新数据集的全部代码与数据、计算得到的描述符总览、配体与实验结果总览，以及基于本文提出的迁移学习方法生成的集成预测结果的交互式版本。