Molecular basis of quantitative structure-properties relationships (QSPR):: A quantum similarity approach
Molecular basis of quantitative structure-properties relationships (QSPR):: A quantum similarity approach
复制标题
DOI：
10.1023/a:1008059505361
复制
发表时间：
1999-05-01
影响因子：
3.5
通讯作者：
Carbó-Dorca, R
中科院分区：
文献类型：
--
作者：
Ponec, R;Amat, L;Carbó-Dorca, R
自从定量结构-性质关系(QSPR)诞生以来，与结构、电子和疏水分子性质相关的经验参数一直被用作确定这些关系的分子描述符。在所有这些参数中，Hammett sigma常数和辛醇-水分配系数的对数logP被大量用于QSPR研究。本文提出了一种新的基于量子相似度量(QSM)的分子描述符，作为这些经验参数的一般替代。这项工作继续了以前关于QSM在QSPR中的应用的分析，在某些情况下引入了分子量子自相似度量(MQS-SM)作为单一的工作参数。MQS-SM作为分子描述符的使用首次从与上述经验参数的相关性中得到证实。用MQS-SM方法研究了一系列取代羧酸的Hammett方程。然后，对于一系列脂肪醇和醋酸酯，用水和辛醇中的密度泛函之间的自相似度量关联了给定分子的logP值。最后，给出了MQS-SM在定量构效关系确定中的应用实例。在所有研究的案例中，MQS-SM似乎是在化学兴趣的一般QSPR应用中使用的优秀的分子描述符。
Since the dawn of quantitative structure-properties relationships (QSPR), empirical parameters related to structural, electronic and hydrophobic molecular properties have been used as molecular descriptors to determine such relationships. Among all these parameters, Hammett sigma constants and the logarithm of the octanol-water partition coefficient, log P, have been massively employed in QSPR studies. In the present paper, a new molecular descriptor, based on quantum similarity measures (QSM), is proposed as a general substitute of these empirical parameters. This work continues previous analyses related to the use of QSM to QSPR, introducing molecular quantum selfsimilarity measures (MQS-SM) as a single working parameter in some cases. The use of MQS-SM as a molecular descriptor is first confirmed from the correlation with the aforementioned empirical parameters. The Hammett equation has been examined using MQS-SM for a series of substituted carboxylic acids. Then, for a series of aliphatic alcohols and acetic acid esters, log P values have been correlated with the self-similarity measure between density functions in water and octanol of a given molecule. And finally, some examples and applications of MQS-SM to determine QSAR are presented. In all studied cases MQS-SM appeared to be excellent molecular descriptors usable in general QSPR applications of chemical interest.