Dataset for Efficient deep learning framework for multiscale simulations in enzymes via linear response approximation

Training data for the three benchmarked QM/MM systems in FIREANN and REANN: SN2 reaction in solution: Cl− + CH3Cl′ → ClCH3 + Cl−, which contains 6 QM atoms; acylation of cyclooxygenase (COX) by aspirin with 26 QM atoms; decomposition of Polyethylene terephthalate (PET) oligomers catalyzed by PETase with 63 QM atoms.用于FIREANN与REANN中三个基准测试量子力学/分子力学（QM/MM）体系的训练数据集：溶液相中的双分子亲核取代（SN2）反应：Cl⁻ + CH3Cl′ → ClCH3 + Cl⁻，该体系包含6个QM原子；阿司匹林对环氧合酶（COX）的酰化反应，该体系含26个QM原子；PETase催化的聚对苯二甲酸乙二酯（PET）低聚物降解反应，该体系含63个QM原子。