Receiver Operating Characteristic Curve Analysis

Since in pharmacophore-based VS only a part of the database molecules matches the model and therefore obtains an alignment score, a limited number of Se and 1-Sp values can be calculated and plotted. Thus, a selective model would result in an ROC curve that starts at the origin and finishes before it reaches the upper right comer where all active and inactive molecules are scored. However, to simplify the explanation of the ROC curve method, we suggest that the model would retrieve and score all database molecules, either because all database molecules are very similar or because partial matching of model features is allowed. In this case, an ideal pharmacophore model will score all actives higher than inactive database molecules. 

---

Wintermark M, Flanders AE, Velthuis B, Meuh R, van Leeuwen M, Goldsher D, Pineda C, Serena J, van der Schaaf I, Waaijer A, Anderson J, Nesbit G, Gabriely L Medina V, Quiles A, Pohlman S, Quist M, Schnyder P, Bogousslavsky J, Dillon WP, Pedraza S (2006) Perfusion-CT assessment of infarct core and penumbra. Receiver operating characteristic curve analysis in 130 patients suspected of acute hemispheric stroke. Stroke 37 979-985. 

Shultz EK. Multivariate receiver-operating characteristic curve analysis Prostate cancer screening as an example. Clin Chem 1995 41 1248-55. 

Wintermark, M., et al., Perfusion-CT assessment of infarct core and penumbra receiver operating characteristic curve analysis in 130 patients suspected of acute hemispheric stroke. Stroke, 2006. 37(4) p. 979-85. 

Kondo H, Kanematsu M, Stiratori Y et al (2001) MR cholangiography with volume rendering receiver operating characteristics curve analysis in patients with choledo-cholithiasis. AJR Am J Roentgenol 176 1183-1189... 

The Receiver Operator Characteristic curve (ROC curve) is a graphical plot of the sensitivity Sn versus false positive rate FPR for a binary classifier system as its discrimination threshold is varied. The ROC curve can also be represented equivalently by plotting the fraction of true positives (TP) versus the fraction of false positives (FP) (Figure C3). ROC analysis provides tools to select possibly optimal classification models. 

Constructing a receiver operating characteristic (ROC) curve, which is a quantitative analysis, it showed that a rapamycin blood concentration of >8 ng/mL was the proper cutoff to define high blood concentration of the drug, and that it was in agreement with the mean rapamycin blood concentration in patients having no restenosis (7.9 ng/mL). 

The recognition accuracy estimation described above faces one very important problem what is the best choice for the threshold value 0 To solve this problem, statistical decision theory is used. ° The basis for this is an analysis of the so-called the Received Operating Characteristic (ROC) curve. By tradition, ROC is plotted as a function of true positive rate TPj TP + FN) (or sensitivity) versus false positive rate FPj TN+FP) (or 1-Specificity) for all possible threshold values 0. Figure 6.5 presents an example of such a ROC curve for the results obtained with our computer program PASS in predicting antineoplastic activity. 

The aim of BPS-MS approach presented here can be summarized by comparing the simulated receiver operator characteristic (ROC) curves shown in Figure 1. Each additional dimension of analysis (a distinct pulse shape, that yields distinct mass spectrum) enlarges the area under the curve and therefore the confidence of the measurement. 

At a high level, the primary objective for statistical analysis of preclinical safety biomarker studies is to assess the association between biomarker levels and histopathological alterations. Although there are many useful and popular statistical methods to analyze this association (i.e., correlation, regression, receiver operating characteristic (ROC) curves), careful thought is needed to conduct a meaningful analysis and yield the most applicable outcomes. 

Although this approach is widespread, it is unsatisfactory for a number of reasons. For one, recovery rates tend not to sufficiently account for early enrichments in sets of retrieved compounds. This deficiency can be partially eliminated using cumulative recall curves, which plot the fraction of actives against the number of compounds retrieved [58, 150]. These curves are similar to the receiver operating characteristic (ROC) curves that have become a popular in compound retrieval studies. Truchon and Bayly [151] have carried out a comprehensive analysis ROC curves and related metrics and have developed an optimal, statistically more robust index, the BEDROC (Boltzmann-enhanced discrimination of ROC) metric. 

To overcome these problems, the operation of the diffusion cell in the transient mode will reveal the contribution of the dead end pores. It is important to obtain the information of these dead end pores as they are usually the pores providing most of the adsorption capacity in the pellet. The principles of the steady state and transient operations of the diffusion cell are very similar to the principles of the time lag presented in the last chapter. The dead end pores are not reflected in the time lag information. Their information must be obtained from the analysis of the transient curve describing the approach of the receiving reservoir s pressure towards steady state. Thus, in order to understand the diffusion characteristic of a pellet, both the steady state and transient operations should be carried out. 

---