Scoring applications

Grosdidier, S. and Fernandez-Recio, J. (2009) Docking and scoring applications to drug discovery in the interactomics era. Expert Opinion on Drug Discovery, 4, 673-686. 

First, one can check whether a randomly compiled test set is within the modeling space, before employing it for PCA/PLS applications. Suppose one has calculated the scores matrix T and the loading matrix P with the help of a training set. Let z be the characteristic vector (that is, the set of independent variables) of an object in a test set. Then, we first must calculate the scores vector of the object (Eq. (14)). 

Like VDA 6.1, AVSQ 94 does not include the requirements of ISO 9001. In this way issues of copyright are overcome, a practice shared by VDA and EAQF but not QS-9000. However, unlike VDA 6.1, AVSQ 94 follows the 20 elements of ISO 9001 with two additional elements, covering financial considerations and product safety. Those questions that go beyond ISO 9001 are marked and as every question is numbered it simplifies the evaluation process. A scoring method is employed to classify organizations in terms of a conformity index. Each question is awarded a point (0, 2.5, 5, 7.5, or 10), where 10 points means full compliance, 7.5 points means minor inadequacies, 5 points means inadequacies in application requiring improvement, 2.5 points means serious inadequacies in application, and 0 points is used for criteria not applied. Unfortunately all questions carry the same weight as no account of the impact of omission on product quality or customer satisfaction is included. 

To demonstrate the excellent correlation (r- = 0.99) between the luminance of the images and molecular diversity, we plotted the luminance values of the map versus the mean similarity values of data sets (Fig. 4-13). From this plot, a scoring scheme for the classification of CSPs from specific to broad application range can be well established Crownpak CR > Pirkle DNBPG > Whelk > Chiralpak AD > Chiralcel OD. 

CDC. 2001. Pesticide applications and field posting. New Jersey Agricultural Experiment Station. Center for Disease Control. Http //search.cdc.gov/search97cgi/s...xt=methvl-Fparathion Sortfield=Score. January 17, 2001. 

Kitchen DB, Decornez H, Furr JR, Bajorath J. Docking and scoring in virtual screening for drug discovery methods and applications. Nat Rev Drug Discov 2004 3(ll) 935-49. 

In their fundamental paper on curve resolution of two-component systems, Lawton and Sylvestre [7] studied a data matrix of spectra recorded during the elution of two constituents. One can decide either to estimate the pure spectra (and derive from them the concentration profiles) or the pure elution profiles (and derive from them the spectra) by factor analysis. Curve resolution, as developed by Lawton and Sylvestre, is based on the evaluation of the scores in the PC-space. Because the scores of the spectra in the PC-space defined by the wavelengths have a clearer structure (e.g. a line or a curve) than the scores of the elution profiles in the PC-space defined by the elution times, curve resolution usually estimates pure spectra. Thereafter, the pure elution profiles are estimated from the estimated pure spectra. Because no information on the specific order of the spectra is used, curve resolution is also applicable when the sequence of the spectra is not in a specific order. 

The application of principal components regression (PCR) to multivariate calibration introduces a new element, viz. data compression through the construction of a small set of new orthogonal components or factors. Henceforth, we will mainly use the term factor rather than component in order to avoid confusion with the chemical components of a mixture. The factors play an intermediary role as regressors in the calibration process. In PCR the factors are obtained as the principal components (PCs) from a principal component analysis (PC A) of the predictor data, i.e. the calibration spectra S (nxp). In Chapters 17 and 31 we saw that any data matrix can be decomposed ( factored ) into a product of (object) score vectors T(nxr) and (variable) loadings P(pxr). The number of columns in T and P is equal to the rank r of the matrix S, usually the smaller of n or p. It is customary and advisable to do this factoring on the data after columncentering. This allows one to write the mean-centered spectra Sq as ... 

The results of such multiple paired comparison tests are usually analyzed with Friedman s rank sum test [4] or with more sophisticated methods, e.g. the one using the Bradley-Terry model [5]. A good introduction to the theory and applications of paired comparison tests is David [6]. Since Friedman s rank sum test is based on less restrictive, ordering assumptions it is a robust alternative to two-way analysis of variance which rests upon the normality assumption. For each panellist (and presentation) the three products are scored, i.e. a product gets a score 1,2 or 3, when it is preferred twice, once or not at all, respectively. The rank scores are summed for each product i. One then tests the hypothesis that this result could be obtained under the null hypothesis that there is no difference between the three products and that the ranks were assigned randomly. Friedman s test statistic for this reads... 

---