Data manipulation and maximum efficient extraction of useful information need a range of
searching, modeling, mathematical, and statistical approaches. Hence, an adequate multivariate
characterization is the first necessary step in investigation and the results are interpreted after multivariate analysis.
Multivariate data analysis is capable of not only large dataset management but also interpret them surely and rapidly.
Application of chemometrics and cheminformatics methods may be useful for design and discovery of new drug
compounds. In this review, we present a variety of information sources on chemometrics, which we consider useful in
different fields of drug design. This review describes exploratory analysis (PCA), classification and multivariate
calibration (PCR, PLS) methods to data analysis. It summarizes the main facts of linear and nonlinear multivariate data
analysis in drug discovery and provides an introduction to manipulation of data in this field. It handles the fundamental
aspects of basic concepts of multivariate methods, principles of projections (PCA and PLS) and introduces the popular
modeling and classification techniques. Enough theory behind these methods, more particularly concerning the
chemometrics tools is included for those with little experience in multivariate data analysis techniques such as PCA, PLS,
SIMCA, etc. We describe each method by avoiding unnecessary equations, and details of calculation algorithms. It
provides a synopsis of the method followed by cases of applications in drug design (i.e., QSAR) and some of the features
for each method.