Compound score profiles

The most commonly adopted approach is to rank compounds according to the total number of alerts. If molecules are to be prioritized with respect to properties such as permeability, solubility, and so forth, the goal is to find the best compromise between these continuous parameters. For this purpose, each property is scored by assigning a dimensionless scale di between zero ( criterion not met ) and one ( target achieved ), with intermediate values being obtained by interpolation between the minimum and maximum for each property. The score profile for each compound is summarized into an overall desirability D by calculating the geometric mean of the individual dt values [67, 68] ... 

In the probabilistic scoring approach, a scoring profile is defined to reflect the profile of properties required for an ideal compound in the context of a project (an example is shown in Figure 8.11). This profile may include simple calculated characteristics, predicted properties, or experimental endpoints. Underlying each of the property criteria is a desirability function that defines the importance of the property to the overall objective of the project and the acceptable compromises. These desirability functions are defined in terms of the impact of the property on the chance of success of the compound a low desirability indicates a low chance of success, or equivalently a high risk, due to the value of the property. Thus, the overall score will reflect the best estimate of the overall chance of success of a compound. As a probability, the overall score will be between zero and one and is multiplicative with respect to the contributions of the individual properties. 

Fig. 15.4 Docking results Typical score profile predicted by docking to 300 snapshots periodically extracted from several simulations of influenza neuraminidase over 30 ns sampling time for a zanamivir and b compound 2, known to bind the open conformation of influenza neuraminidase. The histogram visualizations (c, d) show that best scoring values are obtained with conformations from the ligand-bound simulation. Docking poses (6/acA ) discussed in the text are shown in comparison with native poses of zanamivir green) and compound 2 blue) (e-g)...

Figure 15.5 Scoring profile for selection of compounds intended for an orally administered drug against a peripheral target, based on predicted ADME and physicochemical properties. A desired criterion is defined for...

Figure 15.7 Property profiles and score plots for series 8,11, and 13, derived from the screening library Illustrated In Figure 15.6. The property profiles (top) show the percentage of compounds In the series that meet each property criteria In the scoring profile shown In Figure 15.5 (the bars are In the same order as the properties In the profile). The score plots...

Fig. 18.8 Similarity profile for filtered set of commercially available compounds. 5000 randomly selected compounds from the Available Chemicals Directory that pass the REOS filter were ranked according to their Tanimoto similarity scores (vertical axis) using Daylight fingerprints. 2886 compounds (58%) had similarity scores below 0.85.

It is suggested in this chapter that it is better to use prioritization methods, which utilize a series of soft scores so that compounds with the most interesting profile rank higher and compounds with the least desirable profile rank lower. 

B-score Biological profile score based on potency, selectivity, and toxicity of a compound. 

The D-score is computed using the maximum dissimilarity algorithm of Lajiness (20). This method utilizes a Tanimoto-like similarity measure defined on a 360-bit fragment descriptor used in conjunction with the Cousin/ChemLink system (21). The important feature of this method is that it starts with the selection of a seed compound with subsequent compounds selected based on the maximum diversity relative to all compounds already selected. Thus, the most obvious seed to use in the current scenario is the compound that has the best profile based on the already computed scores. Thus, one needs to compute a preliminary consensus score based on the Q-score and the B-score using weights as defined previously. To summarize this, one needs to... 

In the case of pineapples, the 12 odorants listed in Table 16.7 were dissolved in water in concentrations equal to those determined in the fruit [50]. Then the odour profile of this aroma model was evaluated by a sensory panel in comparison to fresh pineapple juice. The result was a high agreement in the two odour profiles. Fresh, fruity and pineapple-like odour notes scored almost the same intensities in the model as in the juice. Only the sweet aroma note was more intense in the model than in the original sample [50]. In further experiments, the contributions of the six odorants showing the highest OAV (Table 16.7) were evaluated by means of omission tests [9]. The results presented in Table 16.8 show that the omission of 4-hydroxy-2,5-dimethyl-3(2H)-furanone, ethyl 2-methylbutanoate or ethyl 2-methylpropanoate changed the odour so clearly that more than half of the assessors were able to perceive an odour difference between the reduced and the complete aroma model. Therefore, it was concluded that these compounds are the character-impact odorants of fresh pineapple juice. 

WFA starts with the PCA decomposition of the D matrix, giving the product of scores and loadings, TPT. In general, the D matrix will have n components, i.e., rank n. The determination of the location of concentration windows for each component is carried out using EFA (see Figure 11.4b) or other methods. Steps 3 to 5 are the core of the WFA method and should be performed as many times as compounds are present in matrix D to recover the concentration profiles of the C matrix, one at a time. 

The rank of a matrix is a mathematical concept that relates to the number of significant compounds in a dataset, in chemical terms to the number of compounds in a mixture. For example, if there are six compounds in a chromatogram, the rank of the data matrix from the chromatogram should ideally equal 6. However, life is never so simple. What happens is that noise distorts this ideal picture, so even though there may be only six compounds, either it may appear that the rank is 10 or more, or else the apparent rank might even be reduced if the distinction between the profiles for certain compounds are indistinguishable from the noise. If a 15 x 300 X matrix (which may correspond to 15 UV/vis spectra recorded at 1 nm intervals between 201 and 500 nm) has a rank of 6, the scores matrix T has six columns and the loadings matrix P has six rows. 

The dataset in Table 6.2 is of the same size but represents three partially overlapping peaks. The profile (Figure 6.5) appears to be slightly more complex than that for dataset A, and die PC scores plot presented in Figure 6.6 definitely appears to contain more features. Each turning point represents a pure compound, so it appears that there are three compounds, centred at times 9, 13 and 17. In addition, the spectral characteristics... 

Figure 12.1 Multivariate preference mapping of a set of 1064 compounds that were profiled with 11 descriptors (see text for explanation). The upper scores plot shows the projection of the molecules onto the first two components of the principal components analysis. The lower loadings plot illustrates how the descriptors contribute to the positions of the projected compounds in the scores plot.

---