Spectral map analysis

For a detailed description of spectral map analysis (SMA), the reader is referred to Section 31.3.5. The method has been designed specifically for the study of drug-receptor interactions [37,44]. The interpretation of the resulting spectral map is different from that of the usual principal components biplot. The former is symmetric with respect to rows and columns, while the latter is not. In particular, the spectral map displays interactions between compounds and receptors. It shows which compounds are most specific for which receptors (or tests) and vice versa. This property will be illustrated by means of an analysis of data reporting on the binding affinities of various opioid analgesics to various opioid receptors [45,46]. In contrast with the previous approach, this application is not based on extra-thermodynamic properties, but is derived entirely from biological activity spectra. 

Inhibition constants (Kj) of 26 narcotic analgesic agonists and antagonists as obtained in 4 receptor binding tests (EKC, DHM, DADLE and NAL) [45], 

The spectral map shows three distinct poles of specificity. These are respectively the p-receptor (DHM), the 6-receptor (DADLE) and the K-receptor (EKC). The naloxone receptor (NAL) appears to be strongly correlated with the p-receptor (DHM) and, hence, provides little additional information. In spectral map analysis, correlation between variables, as well as similarity between compounds, is evidenced by the proximity of their corresponding symbols. The lines drawn through the three poles of the map represent bipolar axes of contrast. A contrast is defined in this context as a log ratio or, equivalently, as a difference of logarithms. For example, the horizontal axis through the p- and 8-receptors defines the p/6 contrast. Compounds that project on the right side of this axis bind more specifically to the p-receptor, while those that project on the left side possess more 

It appears from the spectral map that the K-receptor is a highly specific receptor which produces strong contrasts in binding affinities of opioid analgesics. The contrast is most evident in ketazocine, ethylketazocine and buprenorphine which possess much more affinity for the K-receptor than for the two others. The contrast is also strong with dihydromorphine, beta-endorphin, an enkephalin analog and two experimental compounds (LY and FK) which have little or no affinity for the K-receptor. 

A Antidiarrhoeal effect B Diuretic effect C Antiptotic effect D Clonidine binding E Idazoxan binding F WB-4101 binding G Geometric mean 

For the compounds and tests represented by the activity spectra shown in Fig. 8.11, a single major contrast allowed the correct classification to be made visually. However, when more than one important contrast is present in the data it becomes difficult, if not impossible, to identify these contrasts by simple inspection of activity spectra. This is where SMA comes in as it is a graphical method for displaying all of the contrasts between the various log ratios in a data table. The SMA process consists of the following steps. 

Row centring of the data (subtraction of the mean activity of a compound in all tests). 

Application of factor analysis to the doubly centred data. 

---

The transformation by log double-centering has received various names among which spectral mapping [13], logarithmic analysis [14], saturated RC association model [15], log-bilinear model [16] and spectral map analysis or SMA for short [17]. 

Each of the three approaches will be applied in this section to the transformed retention times of the 23 chalcones with eight chromatographic elution methods in Table 31.2. The transformation is defined by the successive operations of logarithms, double-centering and global normalization which is typical for the method of spectral map analysis (SMA) ... 

P.J. Lewi, Spectral map analysis. Analysis of contrasts, especially from log-ratios. Chemom. Intell. Lab. Syst., 5 (1989) 105-116. 

According to Andersen [12] early applications of LLM are attributed to the Danish sociologist Rasch in 1963 and to Andersen himself. Later on, the approach has been described under many different names, such as spectral map analysis [13,14] in studies of drug specificity, as logarithmic analysis in the French statistical literature [15] and as the saturated RC association model [16]. The term log-bilinear model has been used by Escoufier and Junca [ 17]. In Chapter 31 on the analysis of measurement tables we have described the method under the name of log double-centred principal components analysis. 

Multivariate methods of data analysis were first applied in chromatography for retention prediction purposes [7. More recently, principal component analysis (PCA), correspondence factor analysis (CFA) and spectral mapping analysis (SMA) have been employed to objectively cla.ssify. stationary phase materials according to the retention... 

Levi, P.J. Spectral map analysis. Eactorial analysis of contrast, especially from log ratios. Chemometr. InteU. Lah. Syst. 1989, 5, 105-116. 

Hamoir, T. Cuaste Sanchez, F. Bourguignon, B. Massart, D.L. Spectral mapping analysis A method for the characterization of stationary phases. J. Chromatogr. Sci. 1994, 32, 488-498. 

Ivaniuc, O. Ivanciuc, T. Cahrol-Bass, D. Balahan, A.T. Com, D.L. Spectral mapping analysis A method for the comparison of weighting schemes for molecular graph descriptors. Application in quantitative structure-retention relationship models for alkylphenols in gas-hquid chromatography. J. Chem. Inf Comput. Sci. 2000, 40, 732-743. 

P. Lewi, Spectral Map Analysis Factorial Analysis of Contrasts, Especially from log ratios, Chemometrics Intelligent Laboratory Systems 5 (1989), 105-116. 

---