The nature and origin of a fundamental quantum QSPR （QQSPR） equation are discussed. In principle, as any molecular structure can be associated to quantum mechanical density functions （DF）, a molecular set can be r...The nature and origin of a fundamental quantum QSPR （QQSPR） equation are discussed. In principle, as any molecular structure can be associated to quantum mechanical density functions （DF）, a molecular set can be reconstructed as a quantum multimolecular polyhedron （QMP）, whose vertices are formed by each molecular DF. According to QQSPR theory, complicated kinds of molecular properties, like biological activity or toxicity, of molecular sets can be calculated via the quantum expectation value of an approximate Hermitian operator, which can be evaluated with the geometrical information contained in the attached QMP via quantum similarity matrices. Practical ways of solving the QQSPR problem from the point of view of QMP geometrical structure are provided. Such a development results into a powerful algorithm, which can be implemented within molecular design as an alternative to the current classical QSPR procedures.展开更多
以物质的电子、空间等结构性质为基础,运用Gaussian98和Cerius2程序包对偶极距(Dipole)、最高占据轨道能量(EHOMO)、最低空轨道能量(ELUMO)、分子总能量(E)、旋转键(Rotlbonds)、最弱的R-NO2键长(R-NO2 bond length,R为C或N)、氢键供体(...以物质的电子、空间等结构性质为基础,运用Gaussian98和Cerius2程序包对偶极距(Dipole)、最高占据轨道能量(EHOMO)、最低空轨道能量(ELUMO)、分子总能量(E)、旋转键(Rotlbonds)、最弱的R-NO2键长(R-NO2 bond length,R为C或N)、氢键供体(Hbond donor)和中点势(Vmid)8种描述符进行了计算,采用Cerius2程序包中的QSPR方法建立了芳香系炸药密度与8种描述符之间的构效关系式,相关系数R为0.909,30个化合物所构成的训练集和15个化合物所构成的预测集预测密度与实测密度之间的平均误差分别为3.33%和2.94%。展开更多
文摘The nature and origin of a fundamental quantum QSPR （QQSPR） equation are discussed. In principle, as any molecular structure can be associated to quantum mechanical density functions （DF）, a molecular set can be reconstructed as a quantum multimolecular polyhedron （QMP）, whose vertices are formed by each molecular DF. According to QQSPR theory, complicated kinds of molecular properties, like biological activity or toxicity, of molecular sets can be calculated via the quantum expectation value of an approximate Hermitian operator, which can be evaluated with the geometrical information contained in the attached QMP via quantum similarity matrices. Practical ways of solving the QQSPR problem from the point of view of QMP geometrical structure are provided. Such a development results into a powerful algorithm, which can be implemented within molecular design as an alternative to the current classical QSPR procedures.
文摘以物质的电子、空间等结构性质为基础,运用Gaussian98和Cerius2程序包对偶极距(Dipole)、最高占据轨道能量(EHOMO)、最低空轨道能量(ELUMO)、分子总能量(E)、旋转键(Rotlbonds)、最弱的R-NO2键长(R-NO2 bond length,R为C或N)、氢键供体(Hbond donor)和中点势(Vmid)8种描述符进行了计算,采用Cerius2程序包中的QSPR方法建立了芳香系炸药密度与8种描述符之间的构效关系式,相关系数R为0.909,30个化合物所构成的训练集和15个化合物所构成的预测集预测密度与实测密度之间的平均误差分别为3.33%和2.94%。
