Premium
Dimensional response of the integrated molecular transform
Author(s) -
King James W.,
Kassel Ronald J.
Publication year - 1991
Publication title -
international journal of quantum chemistry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.484
H-Index - 105
eISSN - 1097-461X
pISSN - 0020-7608
DOI - 10.1002/qua.560400726
Subject(s) - chemistry , wiener index , branching (polymer chemistry) , fourier transform , integral transform , enthalpy , series (stratigraphy) , molecule , computational chemistry , thermodynamics , mathematics , physics , mathematical analysis , combinatorics , graph , organic chemistry , paleontology , biology
The integrated molecular transform provides a unitary chemical structure index that is useful in quantitative structure–activity/property studies. The transform is generated by a Fourier operation on any one of several different matrices of structure representation, e.g., adjacency, bond count, distance between bonded atoms (termed 2‐dimensional [2‐D]), and distance between all atoms in a molecule (termed 3‐dimensional [3‐D]). In this study, the integrated transforms based on bond count and 2‐D and 3‐D structure matrices have been generated and used to correlate an enthalpy function and heats of formation in a series of linear and branched hydrocarbons; these indices were also compared with the 3‐D Wiener number and Randić branching indices for the series. For the linear regression model, the branching and 3‐D Wiener indices correlate the enthalpy function best, but the 2‐D molecular transform is most appropriate for heats of formation. A normalized linear model shows that the 3‐D Wiener number and 2‐D molecular transform are most appropriate for the former data, whereas the latter is most effectively correlated with the 2‐D molecular transform. For reasons not clear, the 3‐D molecular transform is a less suitable index for the noted thermodynamic functions in this series.