About the prediction of molecular properties using the fundamental Quantum QSPR (QQSPR) equation†

Abstract

The present theoretical study analyses the Quantum QSPR fundamental linear equation predictive power. Two main alternative algorithms, among several possible choices, are fully described in an add one and add many basis, while the other possibilities are only sketched out. It is shown that one can also apply the described Quantum QSPR prediction algorithms to parent problems in the framework of empirical QSPR, based on the molecular space framework.

Notes on quantitative structure‐properties relationships (QSPR) part 2: The role of the number of atoms as a molecular descriptor

“…Therefore, QQSPR, even classical QSPR models, obtained by means of the QQSPR-like algorithms derived in part 1, 16 in previous studies 49,53,54 and also in the initial references, [1][2][3][4] seems that cannot be considered at the same level as the descriptor manipulations leading to empirically defined QSPR models, obtained without any causal background.…”

“…This kind of simplification of the fundamental QQSPR equations has been formerly tested in many application examples. [39][40][41][42][43][44][45][46][47][48][49] Due that one can consider the second term in eq. (14) almost a constant in a homogeneous CS fundamental QQSPR treatment, that is:…”

Toward a universal quantum QSPR operator

“…In recent publications, it has been developed the theoretical background, leading into the establishment of a quantum quantitative structure–properties relations (QSPR) equation defined in molecular spaces, resulting from the quantum similarity framework development evolving in time. Recently, a basic concept for this endeavor appears grounded on the description of molecular quantum polyhedra .…”

Centroid origin shift of quantum object sets and molecular point clouds description and element comparisons

“…The procedures developed for this mentioned purpose were applied in a classical molecular descriptor vectorial environment, as a way to show the general application of a quantum QSPR (QQSPR) based methodology to both classical and quantum computational backgrounds. Besides, the cited paper has to be considered at the same time as the first contribution of a series of notes in QSPR and as the last contribution of a long list of previous papers published in various sources [4][5][6][7], studying possible dimensionality paradox free QQSPR mathematical processes dealing with the space of molecules 3 instead of being based upon the space of parameters, the usual working background of classical QSPR. The present contribution has to be considered as a contextual study for further development in such research direction.…”

“…The statistical techniques, employed upon the vectors of the space of descriptors, within the empirical QSPR equation search algorithms, which drastically reduce the dimension of molecular descriptors' space, generate the dimensionality paradox. dimensional quantum molecular density functions (DF), the ultimate descriptors containing all the information about the elements of quantum object sets (QOS) 4 [13,14].…”