Structural correlation analysis

Structural Correlation Analysis of Coordination Compounds and their Polytopal... 

To gain insight into chemometric methods such as correlation analysis, Multiple Linear Regression Analysis, Principal Component Analysis, Principal Component Regression, and Partial Least Squares regression/Projection to Latent Structures... 

It is known that several intermolecular interactions are responsible for cyclodextrin complexation, acting simultaneously. These interactions are separable from one another by quantitative structure-reactivity analysis. Furthermore, correlations obtained by the analysis can be discussed in direct connection with actual interactions already elucidated experimentally for the action site of cyclodextrin. Thus, the results must serve to make the background of the correlation analysis more concrete. 

After the first report on the crystal structure correlation between 2,5-DSP and poly-2,5-DSP crystals, a different crystallographic result was reported on a poly-2,5-DSP crystal (Meyer et al., 1978). It was reconfirmed, however, that the first structural analysis was correct (Nakanishi et al., 1979a). 2,5-DSP, crystallized from benzene solution, is highly photoreactive (a-form), while the same compound, sublimed at a rather high temperature... 

C.J.F. ter Braak, Interpreting canonical correlation analysis through biplots of structure correlations and weights. Psychometrika, 55 (1990) 519-531. 

Molecular Shape Analysis. Once a set of shapes or conformations are generated for a chemical or series of analogs, the usual question is which are similar. Similarity in three dimensions of collections of atoms is very difficult and often subjective. Molecular shape analysis is an attempt to provide a similarity index for molecular structures. The basic approach is to compute the maximum overlap volume of the two molecules by superimposing one onto the other. This is done for all pairs of molecules being considered and this measure, in cubic angstroms, can be used as a parameter for mathematical procedures such as correlation analysis. 

Examples of the application of correlation analysis to diene and polyene data sets are considered below. Both data sets in which the diene or polyene is directly substituted and those in which a phenylene lies between the substituent and diene or polyene group have been considered. In that best of all possible worlds known only to Voltaire s Dr. Pangloss, all data sets have a sufficient number of substituents and cover a wide enough range of substituent electronic demand, steric effect and intermolecular forces to provide a clear, reliable description of structural effects on the property of interest. In the real world this is not often the case. We will therefore try to demonstrate how the maximum amount of information can be extracted from small data sets. 

The utility of small ring conjunctive reagents in total syntheses appears to be very great. The convulated structural correlations between the target and a starting material hamper its ready adaptation. The facility and stereochemical implications make the rethinking of retrosynthetie analysis well worth the effort. To illustrate some of this potential, several applications in the total synthesis of natural and unnatural products are analyzed. 

The emphasis in the foregoing parts of this chapter has been deliberately chemical . We have tried to explore the role of substituent constants in relation to understanding the effect of structure on reaction rates and equilibria, with particular reference to the NO2 group as a substituent. This chemical emphasis will continue in the later parts of the chapter, for NO2 and for the other substituents with which we are concerned, but in the present section there will be a change. In Section H.B brief reference was made to the use of substituent constants in the correlation analysis of spectroscopic data, particularly 19F and 13C substituent chemical shifts and infrared frequencies and intensities. These matters must now be explored in greater detail. 

T. Cserhati, A. Kosa and S. Balogh, Comparison of partial least-square method and canonical correlation analysis in a quantitative structure-retention relationship study. J. Biochem. Biophys. Meth., 36 (1998) 131-141. 

Tomlinson, E. Chromatographic hydrophobic parameters in correlation analysis of structure-activity relationships, J. Chromatogr. A, 113(l) l-45, 1975. 

Stenzel H, Rodicker H. 1991. Results of substance groups and structural groups analysis of petroleum fractions and their application in estimating the correlation for predicting properties. Chemische Technik 43 344-346. 

It Is convenient in correlation analysis to describe a data set as having the general structure X-O-Y In which X Is a variable substituent, Y the active site at which some measurable phenomenon occiirsk and S Is a s)celetal group to which X and Y are bonded. 

---